CVPR 2018 追蹤之論文綱要（修正于2020.08.28）

• 講在前面
• 論文目錄

講在前面

• 論壇很多博客都對(duì)論文做了總結(jié)和分類，但就醫(yī)學(xué)領(lǐng)域而言，對(duì)這些論文的篩選信息顯然需要更加精細(xì)的把控，所以自己對(duì)這979篇的論文做一個(gè)大致從名稱上的篩選，希望能找到些能解決當(dāng)前問題的答案。
• 論文鏈接建議直接Google論文名，比去各種論文或頂會(huì)網(wǎng)站找不知道快捷多少。
• 下面皆為機(jī)器翻譯以方便我第一次篩選，我會(huì)慢慢修正，但現(xiàn)在請(qǐng)結(jié)合。有興趣的可以問我要處理這些論文并自動(dòng)翻譯的腳本。
• Respect！

---

論文目錄

論文概要

1.2D_3D Pose Estimation and Action Recognition Using Multitask Deep Learning 使用多任務(wù)深度學(xué)習(xí)的2D_3D姿勢(shì)估計(jì)和動(dòng)作識(shí)別
2.3D Human Pose Estimation in the Wild by Adversarial Learning 通過(guò)對(duì)抗性學(xué)習(xí)在野外進(jìn)行3D人體姿勢(shì)估計(jì)
3.3D Human Sensing, Action and Emotion Recognition in Robot Assisted Therapy of Children With Autism 孤獨(dú)癥兒童機(jī)器人輔助治療中的3D人體感應(yīng)，動(dòng)作和情感識(shí)別
4.3D Object Detection With Latent Support Surfaces 具有潛在支撐面的3D對(duì)象檢測(cè)
5.3D Pose Estimation and 3D Model Retrieval for Objects in the Wild 野外物體的3D姿勢(shì)估計(jì)和3D模型檢索
6.3D-RCNN: Instance-Level 3D Object Reconstruction via Render-and-Compare 3D-RCNN：通過(guò)渲染和比較重建實(shí)例級(jí)3D對(duì)象
7.3D Registration of Curves and Surfaces Using Local Differential Information 使用局部微分信息進(jìn)行曲線和曲面的3D配準(zhǔn)
8.3D Semantic Segmentation With Submanifold Sparse Convolutional Networks 子流形稀疏卷積網(wǎng)絡(luò)的3D語(yǔ)義分割
9.3D Semantic Trajectory Reconstruction From 3D Pixel Continuum 從3D像素連續(xù)體重建3D語(yǔ)義軌跡
10.4DFAB: A Large Scale 4D Database for Facial Expression Analysis and Biometric Applications 4DFAB：用于面部表情分析和生物識(shí)別應(yīng)用程序的大規(guī)模4D數(shù)據(jù)庫(kù)
11.4D Human Body Correspondences From Panoramic Depth Maps 全景深度圖的4D人體對(duì)應(yīng)
12.A2-RL: Aesthetics Aware Reinforcement Learning for Image Cropping A2-RL：用于圖像裁剪的美學(xué)意識(shí)增強(qiáng)學(xué)習(xí)
13.A Bi-Directional Message Passing Model for Salient Object Detection 顯著目標(biāo)檢測(cè)的雙向消息傳遞模型
14.A Biresolution Spectral Framework for Product Quantization 用于產(chǎn)品量化的雙分辨率光譜框架
15.A Causal And-Or Graph Model for Visibility Fluent Reasoning in Tracking Interacting Objects 交互對(duì)象跟蹤中可視性推理的因果圖模型
16.Accurate and Diverse Sampling of Sequences Based on a “Best of Many” Sample Objective 基于“多個(gè)最佳”樣本目標(biāo)的序列的準(zhǔn)確多樣采樣
17.A Certifiably Globally Optimal Solution to the Non-Minimal Relative Pose Problem 非最小相對(duì)姿勢(shì)問題的可證明的全局最優(yōu)解
18.A Closer Look at Spatiotemporal Convolutions for Action Recognition 近距離觀察時(shí)空卷積的動(dòng)作識(shí)別
19.A Common Framework for Interactive Texture Transfer 交互式紋理傳輸?shù)耐ㄓ每蚣?/strong>
20.A Constrained Deep Neural Network for Ordinal Regression 序數(shù)回歸的約束深度神經(jīng)網(wǎng)絡(luò)
21.Action Sets: Weakly Supervised Action Segmentation Without Ordering Constraints 動(dòng)作集：沒有順序約束的弱監(jiān)督動(dòng)作細(xì)分
22.Active Fixation Control to Predict Saccade Sequences 主動(dòng)注視控制可預(yù)測(cè)掃視序列
23.Actor and Action Video Segmentation From a Sentence 句子中的演員和動(dòng)作視頻分割
24.Actor and Observer: Joint Modeling of First and Third-Person Videos 演員和觀察員：第一人稱和第三人稱視頻的聯(lián)合建模
25.AdaDepth: Unsupervised Content Congruent Adaptation for Depth Estimation AdaDepth：用于深度估計(jì)的無(wú)監(jiān)督內(nèi)容一致適應(yīng)
26.A Deeper Look at Power Normalizations 深入了解功率歸一化
27.Adversarial Complementary Learning for Weakly Supervised Object Localization 弱監(jiān)督對(duì)象定位的對(duì)抗互補(bǔ)學(xué)習(xí)
28.Adversarial Data Programming: Using GANs to Relax the Bottleneck of Curated Labeled Data 對(duì)抗性數(shù)據(jù)編程：使用GAN緩解標(biāo)簽化數(shù)據(jù)的瓶頸
29.Adversarial Feature Augmentation for Unsupervised Domain Adaptation 無(wú)監(jiān)督域自適應(yīng)的對(duì)抗特征增強(qiáng)
30.Adversarially Learned One-Class Classifier for Novelty Detection 對(duì)抗性學(xué)習(xí)的一類分類器，用于新穎性檢測(cè)
31.Adversarially Occluded Samples for Person Re-Identification 對(duì)抗性樣本用于人員重新識(shí)別
32.A Face-to-Face Neural Conversation Model 面對(duì)面的神經(jīng)對(duì)話模型
33.A Fast Resection-Intersection Method for the Known Rotation Problem 已知旋轉(zhuǎn)問題的快速后方交集方法
34.A Generative Adversarial Approach for Zero-Shot Learning From Noisy Texts 一種從嘈雜文本中零接觸學(xué)習(xí)的生成對(duì)抗方法
35.A Hierarchical Generative Model for Eye Image Synthesis and Eye Gaze Estimation 眼睛圖像合成和眼睛注視估計(jì)的分層生成模型
36.A High-Quality Denoising Dataset for Smartphone Cameras 用于智能手機(jī)相機(jī)的高質(zhì)量降噪數(shù)據(jù)集
37.A Hybrid l1-l0 Layer Decomposition Model for Tone Mapping 用于色調(diào)映射的混合l1-l0層分解模型
38.Aligning Infinite-Dimensional Covariance Matrices in Reproducing Kernel Hilbert Spaces for Domain Adaptation 對(duì)齊域希爾伯特希爾伯特空間中的無(wú)限維協(xié)方差矩陣的對(duì)齊
39.Alive Caricature From 2D to 3D 從2D到3D的生動(dòng)漫畫
40.A Low Power, High Throughput, Fully Event-Based Stereo System 低功耗，高吞吐量，完全基于事件的立體聲系統(tǒng)
41.Alternating-Stereo VINS: Observability Analysis and Performance Evaluation 交替立體VINS：可觀察性分析和性能評(píng)估
42.A Memory Network Approach for Story-Based Temporal Summarization of 360deg Videos 基于故事的360deg視頻時(shí)間摘要的記憶網(wǎng)絡(luò)方法
43.A Minimalist Approach to Type-Agnostic Detection of Quadrics in Point Clouds 點(diǎn)云中二次元的類型不可知檢測(cè)的極簡(jiǎn)方法
44.AMNet: Memorability Estimation With Attention AMNet：具有記憶力的評(píng)估
45.Analysis of Hand Segmentation in the Wild 野外手部分割分析
46.Analytical Modeling of Vanishing Points and Curves in Catadioptric Cameras 折反射相機(jī)中消失點(diǎn)和曲線的解析模型
47.Analytic Expressions for Probabilistic Moments of PL-DNN With Gaussian Input 高斯輸入的PL-DNN概率矩的解析表達(dá)式
48.Analyzing Filters Toward Efficient ConvNet 分析面向高效ConvNet的過(guò)濾器
49.An Analysis of Scale Invariance in Object Detection SNIP 目標(biāo)檢測(cè)SNIP中尺度不變性的分析
50.Anatomical Priors in Convolutional Networks for Unsupervised Biomedical Segmentation 卷積網(wǎng)絡(luò)中無(wú)監(jiān)督生物醫(yī)學(xué)分割的解剖先驗(yàn)

論文概要

51.An Efficient and Provable Approach for Mixture Proportion Estimation Using Linear Independence Assumption 使用線性獨(dú)立假設(shè)的混合比例估計(jì)的一種有效且可行的方法
52.An End-to-End TextSpotter With Explicit Alignment and Attention 具有明確對(duì)齊和注意力的端到端TextSpotter
53.A Network Architecture for Point Cloud Classification via Automatic Depth Images Generation 通過(guò)自動(dòng)深度圖像生成進(jìn)行點(diǎn)云分類的網(wǎng)絡(luò)架構(gòu)
54.A Neural Multi-Sequence Alignment TeCHnique (NeuMATCH) 神經(jīng)多序列比對(duì)技術(shù)（NeuMATCH）
55.Anticipating Traffic Accidents With Adaptive Loss and Large-Scale Incident DB 利用自適應(yīng)丟失和大規(guī)模事件數(shù)據(jù)庫(kù)預(yù)測(cè)交通事故
56.An Unsupervised Learning Model for Deformable Medical Image Registration 可變形醫(yī)學(xué)圖像配準(zhǔn)的無(wú)監(jiān)督學(xué)習(xí)模型
57.AON: Towards Arbitrarily-Oriented Text Recognition AON：面向任意方向的文本識(shí)別
58.A Papier-Mache Approach to Learning 3D Surface Generation 學(xué)習(xí)3D曲面生成的Papier-Mache方法
59.A Perceptual Measure for Deep Single Image Camera Calibration 深度單像相機(jī)校準(zhǔn)的感官測(cè)量
60.Aperture Supervision for Monocular Depth Estimation 用于單眼深度估計(jì)的光圈監(jiān)控
61.A PID Controller Approach for Stochastic Optimization of Deep Networks 用于深度網(wǎng)絡(luò)隨機(jī)優(yōu)化的PID控制器方法
62.A Pose-Sensitive Embedding for Person Re-Identification With Expanded Cross Neighborhood Re-Ranking 具有擴(kuò)展的跨鄰域重新排列的姿勢(shì)重新識(shí)別的姿勢(shì)識(shí)別嵌入
63.Appearance-and-Relation Networks for Video Classification 視頻分類的外觀和關(guān)系網(wǎng)絡(luò)
64.A Prior-Less Method for Multi-Face Tracking in Unconstrained Videos 無(wú)約束視頻中多面跟蹤的一種先驗(yàn)減少方法
65.Arbitrary Style Transfer With Deep Feature Reshuffle 任意樣式轉(zhuǎn)移，具有深層功能重組
66.A Revised Underwater Image Formation Model 修訂后的水下成像模型
67.Are You Talking to Me? Reasoned Visual Dialog Generation Through Adversarial Learning 你在跟我講話嗎？通過(guò)對(duì)抗學(xué)習(xí)進(jìn)行合理的視覺對(duì)話生成
68.A Robust Method for Strong Rolling Shutter Effects Correction Using Lines With Automatic Feature Selection 一種具有自動(dòng)特征選擇線的強(qiáng)力滾動(dòng)快門效果校正的魯棒方法
69.Art of Singular Vectors and Universal Adversarial Perturbations 奇異向量和普遍對(duì)抗性攝動(dòng)的藝術(shù)
70.Attend and Interact: Higher-Order Object Interactions for Video Understanding 參加和交互：用于視頻理解的高階對(duì)象交互
71.Attentional ShapeContextNet for Point Cloud Recognition 注意ShapeContextNet用于點(diǎn)云識(shí)別
72.Attention-Aware Compositional Network for Person Re-Identification 用于人員重新識(shí)別的注意感知組成網(wǎng)絡(luò)
73.Attention Clusters: Purely Attention Based Local Feature Integration for Video Classification 注意力集群：基于純粹注意力的視頻分類局部特征集成
74.Attentive Fashion Grammar Network for Fashion Landmark Detection and Clothing Category Classification 專注于時(shí)尚地標(biāo)檢測(cè)和服裝類別分類的時(shí)尚語(yǔ)法網(wǎng)絡(luò)
75.Attentive Generative Adversarial Network for Raindrop Removal From a Single Image 細(xì)心的生成對(duì)抗網(wǎng)絡(luò)，用于從單個(gè)圖像中去除雨滴
76.AttnGAN: Fine-Grained Text to Image Generation With Attentional Generative Adversarial Networks AttnGAN：細(xì)化文本到帶有注意生成對(duì)抗網(wǎng)絡(luò)的圖像生成
77.A Twofold Siamese Network for Real-Time Object Tracking 用于實(shí)時(shí)對(duì)象跟蹤的雙重連體網(wǎng)絡(luò)
78.A Two-Step Disentanglement Method 兩步解纏法
79.Audio to Body Dynamics 音頻到人體動(dòng)力學(xué)
80.Augmented Skeleton Space Transfer for Depth-Based Hand Pose Estimation 基于深度的手部姿勢(shì)估計(jì)的增強(qiáng)骨架空間傳遞
81.Augmenting Crowd-Sourced 3D Reconstructions Using Semantic Detections 使用語(yǔ)義檢測(cè)增強(qiáng)人群源3D重建
82.A Unifying Contrast Maximization Framework for Event Cameras, With Applications to Motion, Depth, and Optical Flow Estimation 用于事件攝像機(jī)的統(tǒng)一對(duì)比度最大化框架，應(yīng)用于運(yùn)動(dòng)，深度和光流估計(jì)
83.Automatic 3D Indoor Scene Modeling From Single Panorama 從單個(gè)全景圖進(jìn)行自動(dòng)3D室內(nèi)場(chǎng)景建模
84.AVA: A Video Dataset of Spatio-Temporally Localized Atomic Visual Actions AVA：時(shí)空局部原子視覺動(dòng)作的視頻數(shù)據(jù)集
85.A Variational U-Net for Conditional Appearance and Shape Generation 用于條件外觀和形狀生成的變體U-網(wǎng)
86.Avatar-Net: Multi-Scale Zero-Shot Style Transfer by Feature Decoration Avatar-Net：通過(guò)特征裝飾進(jìn)行多尺度零射擊樣式轉(zhuǎn)移
87.A Weighted Sparse Sampling and Smoothing Frame Transition Approach for Semantic Fast-Forward First-Person Videos 語(yǔ)義快進(jìn)第一人稱視頻的加權(quán)稀疏采樣和平滑幀過(guò)渡方法
88.Baseline Desensitizing in Translation Averaging 平均翻譯中的基線脫敏
89.Benchmarking 6DOF Outdoor Visual Localization in Changing Conditions 在不斷變化的條件下對(duì)6DOF戶外視覺本地化進(jìn)行基準(zhǔn)測(cè)試
90.Between-Class Learning for Image Classification 課間學(xué)習(xí)進(jìn)行圖像分類
91.Beyond Grobner Bases: Basis Selection for Minimal Solvers Grobner基礎(chǔ)之外：最小求解器的基礎(chǔ)選擇
92.Beyond Holistic Object Recognition: Enriching Image Understanding With Part States 超越整體物體識(shí)別：利用零件狀態(tài)豐富圖像理解
93.Beyond the Pixel-Wise Loss for Topology-Aware Delineation 超越像素明智的拓?fù)涿枋?/strong>
94.Beyond Trade-Off: Accelerate FCN-Based Face Detector With Higher Accuracy 權(quán)衡之外：高精度加速基于FCN的人臉檢測(cè)器
95.Bidirectional Attentive Fusion With Context Gating for Dense Video Captioning 具有上下文門控功能的雙向注意力融合，用于密集視頻字幕
96.Bidirectional Retrieval Made Simple 雙向檢索變得簡(jiǎn)單
97.Bilateral Ordinal Relevance Multi-Instance Regression for Facial Action Unit Intensity Estimation 雙邊序貫相關(guān)性多實(shí)例回歸用于面部動(dòng)作單位強(qiáng)度估計(jì)
98.Blazingly Fast Video Object Segmentation With Pixel-Wise Metric Learning 像素明智的度量學(xué)習(xí)，實(shí)現(xiàn)了驚人的快速視頻對(duì)象分割
99.Blind Predicting Similar Quality Map for Image Quality Assessment 盲預(yù)測(cè)相似質(zhì)量圖進(jìn)行圖像質(zhì)量評(píng)估
100.BlockDrop: Dynamic Inference Paths in Residual Networks BlockDrop：殘差網(wǎng)絡(luò)中的動(dòng)態(tài)推理路徑

論文概要

101.Boosting Adversarial Attacks With Momentum 用動(dòng)量來(lái)增強(qiáng)對(duì)抗性攻擊
102.Boosting Domain Adaptation by Discovering Latent Domains 通過(guò)發(fā)現(xiàn)潛在域來(lái)促進(jìn)域適應(yīng)
103.Boosting Self-Supervised Learning via Knowledge Transfer 通過(guò)知識(shí)轉(zhuǎn)移促進(jìn)自我監(jiān)督學(xué)習(xí)
104.Bootstrapping the Performance of Webly Supervised Semantic Segmentation 引導(dǎo)Webly監(jiān)督的語(yǔ)義分割的性能
105.Bottom-Up and Top-Down Attention for Image Captioning and Visual Question Answering 自下而上和自上而下的注意力，用于圖像字幕和視覺問題解答
106.Boundary Flow: A Siamese Network That Predicts Boundary Motion Without Training on Motion 邊界流：無(wú)需運(yùn)動(dòng)訓(xùn)練就可以預(yù)測(cè)邊界運(yùn)動(dòng)的暹羅網(wǎng)絡(luò)
107.BPGrad: Towards Global Optimality in Deep Learning via Branch and Pruning BPGrad：通過(guò)分支和修剪在深度學(xué)習(xí)中實(shí)現(xiàn)全球最優(yōu)
108.Burst Denoising With Kernel Prediction Networks 內(nèi)核預(yù)測(cè)網(wǎng)絡(luò)進(jìn)行突發(fā)去噪
109.Camera Pose Estimation With Unknown Principal Point 主點(diǎn)未知的相機(jī)姿態(tài)估計(jì)
110.Camera Style Adaptation for Person Re-Identification 用于重新識(shí)別人的相機(jī)樣式適應(yīng)
111.Can Spatiotemporal 3D CNNs Retrace the History of 2D CNNs and ImageNet? 時(shí)空3D CNN是否可以追溯2D CNN和ImageNet的歷史？
112.CarFusion: Combining Point Tracking and Part Detection for Dynamic 3D Reconstruction of Vehicles CarFusion：結(jié)合點(diǎn)跟蹤和零件檢測(cè)用于車輛的動(dòng)態(tài)3D重構(gòu)
113.CartoonGAN: Generative Adversarial Networks for Photo Cartoonization CartoonGAN：用于照片卡通化的生成對(duì)抗網(wǎng)絡(luò)
114.Cascaded Pyramid Network for Multi-Person Pose Estimation 用于多人姿勢(shì)估計(jì)的級(jí)聯(lián)金字塔網(wǎng)絡(luò)
115.Cascade R-CNN: Delving Into High Quality Object Detection 級(jí)聯(lián)R-CNN：深入研究高質(zhì)量目標(biāo)檢測(cè)
116.Categorizing Concepts With Basic Level for Vision-to-Language 將基本概念歸類為視覺到語(yǔ)言
117.CBMV: A Coalesced Bidirectional Matching Volume for Disparity Estimation CBMV：用于視差估計(jì)的合并雙向匹配量
118.Classification-Driven Dynamic Image Enhancement 分類驅(qū)動(dòng)的動(dòng)態(tài)圖像增強(qiáng)
119.Classifier Learning With Prior Probabilities for Facial Action Unit Recognition 具有面部動(dòng)作單元識(shí)別先驗(yàn)概率的分類器學(xué)習(xí)
120.ClcNet: Improving the Efficiency of Convolutional Neural Network Using Channel Local Convolutions ClcNet：使用通道局部卷積提高卷積神經(jīng)網(wǎng)絡(luò)的效率
121.CleanNet: Transfer Learning for Scalable Image Classifier Training With Label Noise CleanNet：帶標(biāo)簽噪聲的可擴(kuò)展圖像分類器培訓(xùn)的轉(zhuǎn)移學(xué)習(xí)
122.CLEAR: Cumulative LEARning for One-Shot One-Class Image Recognition 清除：一鍵式一類圖像識(shí)別的累積學(xué)習(xí)
123.Clinical Skin Lesion Diagnosis Using Representations Inspired by Dermatologist Criteria 使用皮膚科醫(yī)生標(biāo)準(zhǔn)啟發(fā)的表征進(jìn)行臨床皮膚病變?cè)\斷
124.CLIP-Q: Deep Network Compression Learning by In-Parallel Pruning-Quantization CLIP-Q：通過(guò)并行修剪量化進(jìn)行深度網(wǎng)絡(luò)壓縮學(xué)習(xí)
125.ClusterNet: Detecting Small Objects in Large Scenes by Exploiting Spatio-Temporal Information ClusterNet：通過(guò)利用時(shí)空信息檢測(cè)大型場(chǎng)景中的小物體
126.CNN Based Learning Using Reflection and Retinex Models for Intrinsic Image Decomposition 基于CNN的使用反射和Retinex模型進(jìn)行內(nèi)在圖像分解的學(xué)習(xí)
127.CNN Driven Sparse Multi-Level B-Spline Image Registration CNN驅(qū)動(dòng)的稀疏多級(jí)B樣條圖像配準(zhǔn)
128.CNN in MRF: Video Object Segmentation via Inference in a CNN-Based Higher-Order Spatio-Temporal MRF MRF中的CNN：在基于CNN的高階時(shí)空MRF中通過(guò)推理進(jìn)行視頻對(duì)象分割
129.COCO-Stuff: Thing and Stuff Classes in Context COCO-Stuff：上下文中的事物和事物類
130.CodeSLAM – Learning a Compact, Optimisable Representation for Dense Visual SLAM CodeSLAM-學(xué)習(xí)密集Visual SLAM的緊湊，可優(yōu)化表示形式
131.Coding Kendall’s Shape Trajectories for 3D Action Recognition 編碼Kendall的形狀軌跡以進(jìn)行3D動(dòng)作識(shí)別
132.Collaborative and Adversarial Network for Unsupervised Domain Adaptation 無(wú)監(jiān)督域自適應(yīng)的協(xié)作和對(duì)抗網(wǎng)絡(luò)
133.Compare and Contrast: Learning Prominent Visual Differences 比較和對(duì)比：學(xué)習(xí)明顯的視覺差異
134.Compassionately Conservative Balanced Cuts for Image Segmentation 慷慨保守的平衡切割用于圖像分割
135.Compressed Video Action Recognition 壓縮視頻動(dòng)作識(shí)別
136.CondenseNet: An Efficient DenseNet Using Learned Group Convolutions CondenseNet：使用學(xué)習(xí)的組卷積的高效DenseNet
137.Conditional Generative Adversarial Network for Structured Domain Adaptation 結(jié)構(gòu)化領(lǐng)域適應(yīng)的條件生成對(duì)抗網(wǎng)絡(luò)
138.Conditional Image-to-Image Translation 有條件的圖像到圖像翻譯
139.Conditional Probability Models for Deep Image Compression 深度圖像壓縮的條件概率模型
140.Connecting Pixels to Privacy and Utility: Automatic Redaction of Private Information in Images 將像素連接到隱私和實(shí)用程序：自動(dòng)刪除圖像中的私人信息
141.Consensus Maximization for Semantic Region Correspondences 語(yǔ)義區(qū)域?qū)?yīng)的共識(shí)最大化
142.Content-Sensitive Supervoxels via Uniform Tessellations on Video Manifolds 通過(guò)視頻流形上的統(tǒng)一鑲嵌來(lái)對(duì)內(nèi)容敏感的超級(jí)體素
143.Context-Aware Deep Feature Compression for High-Speed Visual Tracking 用于高速視覺跟蹤的上下文感知深度特征壓縮
144.Context-Aware Synthesis for Video Frame Interpolation 視頻幀插值的上下文感知綜合
145.Context Contrasted Feature and Gated Multi-Scale Aggregation for Scene Segmentation 用于場(chǎng)景分割的上下文對(duì)比特征和門控多尺度聚合
146.Context Embedding Networks 上下文嵌入網(wǎng)絡(luò)
147.Context Encoding for Semantic Segmentation 用于語(yǔ)義分割的上下文編碼
148.Continuous Relaxation of MAP Inference: A Nonconvex Perspective MAP推理的連續(xù)松弛：非凸視角
149.Controllable Video Generation With Sparse Trajectories 具有稀疏軌跡的可控視頻生成
150.Convolutional Image Captioning 卷積圖像字幕

論文概要

151.Convolutional Neural Networks With Alternately Updated Clique 具有交替更新的派系的卷積神經(jīng)網(wǎng)絡(luò)
152.Convolutional Sequence to Sequence Model for Human Dynamics 卷積序列到人類動(dòng)力學(xué)序列模型
153.Correlation Tracking via Joint Discrimination and Reliability Learning 通過(guò)聯(lián)合鑒別和可靠性學(xué)習(xí)進(jìn)行關(guān)聯(lián)跟蹤
154.CosFace: Large Margin Cosine Loss for Deep Face Recognition CosFace：用于識(shí)別深臉的大余弦余弦損失
155.Coupled End-to-End Transfer Learning With Generalized Fisher Information 端到端遷移學(xué)習(xí)與廣義Fisher信息相結(jié)合
156.Crafting a Toolchain for Image Restoration by Deep Reinforcement Learning 通過(guò)深度強(qiáng)化學(xué)習(xí)制作用于圖像還原的工具鏈
157.Creating Capsule Wardrobes From Fashion Images 從時(shí)尚形象創(chuàng)建膠囊衣柜
158.Cross-Dataset Adaptation for Visual Question Answering 跨數(shù)據(jù)集自適應(yīng)以解決視覺問題
159.Cross-Domain Self-Supervised Multi-Task Feature Learning Using Synthetic Imagery 使用合成影像的跨域自我監(jiān)督多任務(wù)特征學(xué)習(xí)
160.Cross-Domain Weakly-Supervised Object Detection Through Progressive Domain Adaptation 通過(guò)漸進(jìn)域自適應(yīng)進(jìn)行跨域弱監(jiān)督對(duì)象檢測(cè)
161.Cross-Modal Deep Variational Hand Pose Estimation 跨模態(tài)深度變化手姿勢(shì)估計(jì)
162.Cross-View Image Synthesis Using Conditional GANs 使用條件GAN的跨視圖圖像合成
163.Crowd Counting via Adversarial Cross-Scale Consistency Pursuit 通過(guò)對(duì)抗性跨尺度一致性追求進(jìn)行人群計(jì)數(shù)
164.Crowd Counting With Deep Negative Correlation Learning 深度負(fù)相關(guān)學(xué)習(xí)的人群計(jì)數(shù)
165.CRRN: Multi-Scale Guided Concurrent Reflection Removal Network CRRN：多尺度引導(dǎo)并發(fā)反射去除網(wǎng)絡(luò)
166.CSGNet: Neural Shape Parser for Constructive Solid Geometry CSGNet：用于構(gòu)造實(shí)體幾何的神經(jīng)形狀解析器
167.CSRNet: Dilated Convolutional Neural Networks for Understanding the Highly Congested Scenes CSRNet：擴(kuò)展卷積神經(jīng)網(wǎng)絡(luò)，用于了解高度擁擠的場(chǎng)景
168.Cube Padding for Weakly-Supervised Saliency Prediction in 360deg Videos 360度視頻中弱監(jiān)督顯著性預(yù)測(cè)的多維數(shù)據(jù)集填充
169.Curve Reconstruction via the Global Statistics of Natural Curves 通過(guò)自然曲線的整體統(tǒng)計(jì)量重建曲線
170.Customized Image Narrative Generation via Interactive Visual Question Generation and Answering 通過(guò)交互式視覺問題生成和回答定制的圖像敘事生成
171.CVM-Net: Cross-View Matching Network for Image-Based Ground-to-Aerial Geo-Localization CVM-Net：用于基于圖像的地對(duì)空地理定位的跨視圖匹配網(wǎng)絡(luò)
172.DA-GAN: Instance-Level Image Translation by Deep Attention Generative Adversarial Networks DA-GAN：深度注意生成對(duì)抗網(wǎng)絡(luò)的實(shí)例級(jí)圖像翻譯
173.Data Distillation: Towards Omni-Supervised Learning 數(shù)據(jù)提煉：走向全監(jiān)督學(xué)習(xí)
174.DeblurGAN: Blind Motion Deblurring Using Conditional Adversarial Networks DeblurGAN：使用條件對(duì)抗網(wǎng)絡(luò)進(jìn)行盲運(yùn)動(dòng)去模糊
175.DecideNet: Counting Varying Density Crowds Through Attention Guided Detection and Density Estimation DecideNet：通過(guò)注意力指導(dǎo)的檢測(cè)和密度估計(jì)計(jì)算不同的密度人群
176.Decorrelated Batch Normalization 裝飾相關(guān)的批次標(biāo)準(zhǔn)化
177.Decoupled Networks 解耦網(wǎng)絡(luò)
178.Deep Adversarial Metric Learning 深度對(duì)抗度量學(xué)習(xí)
179.Deep Adversarial Subspace Clustering 深度對(duì)抗子空間聚類
180.Deep Back-Projection Networks for Super-Resolution 深度背投網(wǎng)絡(luò)可實(shí)現(xiàn)超高分辨率
181.Deep Cauchy Hashing for Hamming Space Retrieval 深層柯西散列用于漢明空間檢索
182.Deep Cocktail Network: Multi-Source Unsupervised Domain Adaptation With Category Shift 深度雞尾酒網(wǎng)絡(luò)：具有類別轉(zhuǎn)移的多源無(wú)監(jiān)督域自適應(yīng)
183.Deep Cost-Sensitive and Order-Preserving Feature Learning for Cross-Population Age Estimation 用于跨人口年齡估計(jì)的深度成本敏感和順序保留特征學(xué)習(xí)
184.Deep Cross-Media Knowledge Transfer 深度跨媒體知識(shí)轉(zhuǎn)移
185.Deep Density Clustering of Unconstrained Faces 無(wú)約束面孔的深度密度聚類
186.Deep Depth Completion of a Single RGB-D Image 單個(gè)RGB-D圖像的深度完成
187.Deep Diffeomorphic Transformer Networks 深微形變壓器網(wǎng)絡(luò)
188.Deep End-to-End Time-of-Flight Imaging 深度端到端飛行時(shí)間成像
189.Deep Extreme Cut: From Extreme Points to Object Segmentation 深度極限切割：從極限點(diǎn)到對(duì)象分割
190.Deep Face Detector Adaptation Without Negative Transfer or Catastrophic Forgetting 無(wú)需負(fù)遷移或?yàn)?zāi)難性遺忘的深臉檢測(cè)器自適應(yīng)
191.Deep Group-Shuffling Random Walk for Person Re-Identification 用于人員重新識(shí)別的深度群混洗隨機(jī)游走
192.Deep Hashing via Discrepancy Minimization 通過(guò)差異最小化進(jìn)行深度哈希
193.Deep Image Prior 深度圖像先驗(yàn)
194.Deep Layer Aggregation 深層聚合
195.Deep Learning of Graph Matching 圖匹配的深度學(xué)習(xí)
196.Deep Learning Under Privileged Information Using Heteroscedastic Dropout 使用異方差輟學(xué)在特權(quán)信息下進(jìn)行深度學(xué)習(xí)
197.Deep Lesion Graphs in the Wild: Relationship Learning and Organization of Significant Radiology Image Findings in a Diverse Large-Scale Lesion Database 在野外的深部病變圖：關(guān)系學(xué)習(xí)和在大量大型病變數(shù)據(jù)庫(kù)中的重要放射圖像發(fā)現(xiàn)的組織
198.Deeply Learned Filter Response Functions for Hyperspectral Reconstruction 深度學(xué)習(xí)的濾波器響應(yīng)函數(shù)，用于高光譜重建
199.Deep Marching Cubes: Learning Explicit Surface Representations 深入進(jìn)行中的立方體：學(xué)習(xí)明確的表面表示
200.Deep Material-Aware Cross-Spectral Stereo Matching 深度材料感知跨譜立體匹配

論文概要

201.Deep Mutual Learning 深度相互學(xué)習(xí)
202.DeepMVS: Learning Multi-View Stereopsis DeepMVS：學(xué)習(xí)多視圖立體視覺
203.Deep Ordinal Regression Network for Monocular Depth Estimation 用于單眼深度估計(jì)的深度序數(shù)回歸網(wǎng)絡(luò)
204.Deep Parametric Continuous Convolutional Neural Networks 深參量連續(xù)卷積神經(jīng)網(wǎng)絡(luò)
205.Deep Photo Enhancer: Unpaired Learning for Image Enhancement From Photographs With GANs 深度照片增強(qiáng)器：使用GAN從照片中進(jìn)行成對(duì)學(xué)習(xí)的圖像增強(qiáng)
206.Deep Progressive Reinforcement Learning for Skeleton-Based Action Recognition 基于骨骼的動(dòng)作識(shí)別的深度漸進(jìn)強(qiáng)化學(xué)習(xí)
207.Deep Regression Forests for Age Estimation 深回歸森林的年齡估算
208.Deep Reinforcement Learning of Region Proposal Networks for Object Detection 用于對(duì)象檢測(cè)的區(qū)域提議網(wǎng)絡(luò)的深度強(qiáng)化學(xué)習(xí)
209.Deep Semantic Face Deblurring 深層語(yǔ)義去模糊
210.Deep Sparse Coding for Invariant Multimodal Halle Berry Neurons 不變多模態(tài)哈莉·貝瑞神經(jīng)元的深度稀疏編碼
211.Deep Spatial Feature Reconstruction for Partial Person Re-Identification: Alignment-Free Approach 用于部分人員重新識(shí)別的深度空間特征重建：無(wú)路線方法
212.Deep Spatio-Temporal Random Fields for Efficient Video Segmentation 深度時(shí)空隨機(jī)場(chǎng)，用于有效的視頻分割
213.Deep Texture Manifold for Ground Terrain Recognition 用于地面地形識(shí)別的深紋理流形
214.Deep Unsupervised Saliency Detection: A Multiple Noisy Labeling Perspective 深度無(wú)監(jiān)督的顯著性檢測(cè)：多重噪聲標(biāo)記
215.Deep Video Super-Resolution Network Using Dynamic Upsampling Filters Without Explicit Motion Compensation 深度視頻超分辨率網(wǎng)絡(luò)，使用動(dòng)態(tài)上采樣濾波器，無(wú)需顯式運(yùn)動(dòng)補(bǔ)償
216.DeepVoting: A Robust and Explainable Deep Network for Semantic Part Detection Under Partial Occlusion DeepVoting：在部分遮擋下用于語(yǔ)義部分檢測(cè)的強(qiáng)大且可解釋的深度網(wǎng)絡(luò)
217.Defense Against Adversarial Attacks Using High-Level Representation Guided Denoiser 使用高級(jí)表示制導(dǎo)的降噪器防御對(duì)抗攻擊
218.Defense Against Universal Adversarial Perturbations 防御普遍的對(duì)抗性干擾
219.Deflecting Adversarial Attacks With Pixel Deflection 通過(guò)像素偏轉(zhuǎn)來(lái)對(duì)抗對(duì)手的攻擊
220.Defocus Blur Detection via Multi-Stream Bottom-Top-Bottom Fully Convolutional Network 通過(guò)多流底部-頂部-底部完全卷積網(wǎng)絡(luò)進(jìn)行散焦模糊檢測(cè)
221.Deformable GANs for Pose-Based Human Image Generation 用于基于姿勢(shì)的人體圖像生成的可變形GAN
222.Deformable Shape Completion With Graph Convolutional Autoencoders 圖卷積自動(dòng)編碼器的可變形形狀完成
223.Deformation Aware Image Compression 變形感知圖像壓縮
224.DeLS-3D: Deep Localization and Segmentation With a 3D Semantic Map DeLS-3D：具有3D語(yǔ)義圖的深度定位和細(xì)分
225.Demo2Vec: Reasoning Object Affordances From Online Videos Demo2Vec：從在線視頻中推理出對(duì)象客流
226.Dense 3D Regression for Hand Pose Estimation 手勢(shì)姿勢(shì)估計(jì)的密集3D回歸
227.DenseASPP for Semantic Segmentation in Street Scenes DenseASPP用于街道場(chǎng)景中的語(yǔ)義分割
228.Dense Decoder Shortcut Connections for Single-Pass Semantic Segmentation 用于單遍語(yǔ)義分割的密集解碼器快捷方式連接
229.Densely Connected Pyramid Dehazing Network 密集連接的金字塔除霧網(wǎng)絡(luò)
230.DensePose: Dense Human Pose Estimation in the Wild DensePose：野外的密集人體姿勢(shì)估計(jì)
231.Density Adaptive Point Set Registration 密度自適應(yīng)點(diǎn)集配準(zhǔn)
232.Density-Aware Single Image De-Raining Using a Multi-Stream Dense Network 使用多流密集網(wǎng)絡(luò)的密度感知單圖像降噪
233.Depth and Transient Imaging With Compressive SPAD Array Cameras 壓縮SPAD陣列攝像機(jī)的深度和瞬態(tài)成像
234.Depth-Aware Stereo Video Retargeting 深度感知立體聲視頻重定向
235.Depth-Based 3D Hand Pose Estimation: From Current Achievements to Future Goals 基于深度的3D手勢(shì)估計(jì)：從當(dāng)前成就到未來(lái)目標(biāo)
236.Detach and Adapt: Learning Cross-Domain Disentangled Deep Representation 分離和適應(yīng)：學(xué)習(xí)跨域解纏結(jié)的深度表示
237.Detail-Preserving Pooling in Deep Networks 深度網(wǎng)絡(luò)中保留細(xì)節(jié)的池
238.Detect-and-Track: Efficient Pose Estimation in Videos 檢測(cè)并跟蹤：視頻中的有效姿勢(shì)估計(jì)
239.Detect Globally, Refine Locally: A Novel Approach to Saliency Detection 全局檢測(cè)，局部?jī)?yōu)化：顯著性檢測(cè)的新方法
240.Detecting and Recognizing Human-Object Interactions 檢測(cè)和識(shí)別人與物體的相互作用
241.Differential Attention for Visual Question Answering 視覺問答中的注意差異
242.Dimensionality’s Blessing: Clustering Images by Underlying Distribution 維數(shù)的祝福：通過(guò)基礎(chǔ)分布將圖像聚類
243.Direction-Aware Spatial Context Features for Shadow Detection 用于陰影檢測(cè)的方向感知空間上下文功能
244.Direct Shape Regression Networks for End-to-End Face Alignment 直接形狀回歸網(wǎng)絡(luò)用于端對(duì)端的面對(duì)齊
245.Discovering Point Lights With Intensity Distance Fields 發(fā)現(xiàn)具有強(qiáng)度距離場(chǎng)的點(diǎn)光源
246.Discrete-Continuous ADMM for Transductive Inference in Higher-Order MRFs 離散連續(xù)ADMM用于高階MRF中的轉(zhuǎn)導(dǎo)推理
247.Discriminability Objective for Training Descriptive Captions 訓(xùn)練描述性字幕的可分辨性目標(biāo)
248.Discriminative Learning of Latent Features for Zero-Shot Recognition 零射擊識(shí)別的潛在特征的判別學(xué)習(xí)
249.Disentangled Person Image Generation 糾纏人圖像生成
250.Disentangling 3D Pose in a Dendritic CNN for Unconstrained 2D Face Alignment 解開樹枝狀CNN中的3D姿勢(shì)以實(shí)現(xiàn)不受約束的2D面部對(duì)齊

論文概要

251.Disentangling Factors of Variation by Mixing Them 通過(guò)混合將變量分解開來(lái)
252.Disentangling Features in 3D Face Shapes for Joint Face Reconstruction and Recognition 解開3D人臉形狀的特征以進(jìn)行聯(lián)合人臉重建和識(shí)別
253.Disentangling Structure and Aesthetics for Style-Aware Image Completion 解開結(jié)構(gòu)和美學(xué)的風(fēng)格感知圖像完成
254.Distort-and-Recover: Color Enhancement Using Deep Reinforcement Learning 失真與恢復(fù)：使用深度強(qiáng)化學(xué)習(xí)增強(qiáng)色彩
255.Distributable Consistent Multi-Object Matching 可分配一致的多對(duì)象匹配
256.DiverseNet: When One Right Answer Is Not Enough DiverseNet：當(dāng)一個(gè)正確的答案還不夠時(shí)
257.Diversity Regularized Spatiotemporal Attention for Video-Based Person Re-Identification 基于視頻的人員重新識(shí)別的多樣性正則化時(shí)空注意
258.Divide and Conquer for Full-Resolution Light Field Deblurring 分立制勝，實(shí)現(xiàn)全分辨率光場(chǎng)去模糊
259.Divide and Grow: Capturing Huge Diversity in Crowd Images With Incrementally Growing CNN 分而成長(zhǎng)：隨著CNN的不斷增長(zhǎng)，捕捉人群圖像中的巨大多樣性
260.Document Enhancement Using Visibility Detection 使用可見性檢測(cè)增強(qiáng)文檔
261.DocUNet: Document Image Unwarping via a Stacked U-Net DocUNet：文檔圖像通過(guò)堆疊的U-Net變形
262.Domain Adaptive Faster R-CNN for Object Detection in the Wild 用于野外目標(biāo)檢測(cè)的域自適應(yīng)快速R-CNN
263.Domain Generalization With Adversarial Feature Learning 具有對(duì)抗性特征學(xué)習(xí)的領(lǐng)域概括
264.Don’t Just Assume; Look and Answer: Overcoming Priors for Visual Question Answering 不要只是假設(shè)；外觀和答案：克服視覺提問的先驗(yàn)
265.DOTA: A Large-Scale Dataset for Object Detection in Aerial Images DOTA：用于航空?qǐng)D像中目標(biāo)檢測(cè)的大規(guī)模數(shù)據(jù)集
266.DoubleFusion: Real-Time Capture of Human Performances With Inner Body Shapes From a Single Depth Sensor DoubleFusion：通過(guò)單個(gè)深度傳感器實(shí)時(shí)捕獲具有人體形狀的人體表演
267.DS*: Tighter Lifting-Free Convex Relaxations for Quadratic Matching Problems *DS ：針對(duì)二次匹配問題的更緊的免提凸松弛
268.Dual Attention Matching Network for Context-Aware Feature Sequence Based Person Re-Identification 用于基于上下文感知特征序列的人員重新識(shí)別的雙注意匹配網(wǎng)絡(luò)
269.Dual Skipping Networks 雙跳網(wǎng)
270.Duplex Generative Adversarial Network for Unsupervised Domain Adaptation 用于無(wú)監(jiān)督域自適應(yīng)的雙工生成對(duì)抗網(wǎng)絡(luò)
271.DVQA: Understanding Data Visualizations via Question Answering DVQA：通過(guò)問答理解數(shù)據(jù)可視化
272.Dynamic Feature Learning for Partial Face Recognition 動(dòng)態(tài)特征學(xué)習(xí)用于部分人臉識(shí)別
273.Dynamic Few-Shot Visual Learning Without Forgetting 無(wú)需忘記的動(dòng)態(tài)少量視覺學(xué)習(xí)
274.Dynamic Graph Generation Network: Generating Relational Knowledge From Diagrams 動(dòng)態(tài)圖生成網(wǎng)絡(luò)：從圖生成關(guān)系知識(shí)
275.Dynamic Scene Deblurring Using Spatially Variant Recurrent Neural Networks 使用空間變異遞歸神經(jīng)網(wǎng)絡(luò)進(jìn)行動(dòng)態(tài)場(chǎng)景去模糊
276.Dynamic-Structured Semantic Propagation Network 動(dòng)態(tài)結(jié)構(gòu)的語(yǔ)義傳播網(wǎng)絡(luò)
277.Dynamic Video Segmentation Network 動(dòng)態(tài)視頻分割網(wǎng)絡(luò)
278.Dynamic Zoom-In Network for Fast Object Detection in Large Images 動(dòng)態(tài)放大網(wǎng)絡(luò)，可快速檢測(cè)大圖像中的物體
279.Easy Identification From Better Constraints: Multi-Shot Person Re-Identification From Reference Constraints 從更好的約束中輕松識(shí)別：從參考約束中進(jìn)行多次連發(fā)人員重新識(shí)別
280.Edit Probability for Scene Text Recognition 編輯場(chǎng)景文本識(shí)別的概率
281.Efficient and Deep Person Re-Identification Using Multi-Level Similarity 使用多級(jí)相似性進(jìn)行有效的深度人員重新識(shí)別
282.Efficient Diverse Ensemble for Discriminative Co-Tracking 高效的多元化集合體，可進(jìn)行區(qū)分式協(xié)同跟蹤
283.Efficient Interactive Annotation of Segmentation Datasets With Polygon-RNN++ 使用Polygon-RNN ++的分段數(shù)據(jù)集的高效交互式注釋
284.Efficient Large-Scale Approximate Nearest Neighbor Search on OpenCL FPGA 在OpenCL FPGA上進(jìn)行高效的大規(guī)模近似最近鄰居搜索
285.Efficient Optimization for Rank-Based Loss Functions 基于等級(jí)的損失函數(shù)的有效優(yōu)化
286.Efficient Parametrization of Multi-Domain Deep Neural Networks 多域深度神經(jīng)網(wǎng)絡(luò)的高效參數(shù)化
287.Efficient, Sparse Representation of Manifold Distance Matrices for Classical Scaling 流形距離矩陣的有效，稀疏表示
288.Efficient Subpixel Refinement With Symbolic Linear Predictors 使用符號(hào)線性預(yù)測(cè)器進(jìn)行有效的亞像素細(xì)化
289.Efficient Video Object Segmentation via Network Modulation 通過(guò)網(wǎng)絡(luò)調(diào)制進(jìn)行有效的視頻對(duì)象分割
290.Egocentric Activity Recognition on a Budget 預(yù)算中的自我中心活動(dòng)識(shí)別
291.Egocentric Basketball Motion Planning From a Single First-Person Image 從單個(gè)第一人稱圖像進(jìn)行以自我為中心的籃球運(yùn)動(dòng)計(jì)劃
292.Eliminating Background-Bias for Robust Person Re-Identification 消除背景偏見，進(jìn)行穩(wěn)健的人員重新識(shí)別
293.Embodied Question Answering 具體問題解答
294.Emotional Attention: A Study of Image Sentiment and Visual Attention 情緒注意：圖像情感和視覺注意的研究
295.Empirical Study of the Topology and Geometry of Deep Networks 深度網(wǎng)絡(luò)拓?fù)浜蛶缀蔚膶?shí)證研究
296.Encoding Crowd Interaction With Deep Neural Network for Pedestrian Trajectory Prediction 用深度神經(jīng)網(wǎng)絡(luò)編碼人群交互作用以預(yù)測(cè)行人軌跡
297.End-to-End Convolutional Semantic Embeddings 端到端卷積語(yǔ)義嵌入
298.End-to-End Deep Kronecker-Product Matching for Person Re-Identification 端到端深度Kronecker產(chǎn)品匹配以重新識(shí)別人
299.End-to-End Dense Video Captioning With Masked Transformer 帶屏蔽變壓器的端到端密集視頻字幕
300.End-to-End Flow Correlation Tracking With Spatial-Temporal Attention 時(shí)空注意的端到端流相關(guān)跟蹤

論文概要

301.End-to-End Learning of Keypoint Detector and Descriptor for Pose Invariant 3D Matching 姿勢(shì)不變3D匹配的關(guān)鍵點(diǎn)檢測(cè)器和描述符的端到端學(xué)習(xí)
302.End-to-End Learning of Motion Representation for Video Understanding 端到端學(xué)習(xí)運(yùn)動(dòng)表示以了解視頻
303.End-to-End Recovery of Human Shape and Pose 人體形狀和姿勢(shì)的端到端恢復(fù)
304.End-to-End Weakly-Supervised Semantic Alignment 端到端弱監(jiān)督的語(yǔ)義對(duì)齊
305.Enhancing the Spatial Resolution of Stereo Images Using a Parallax Prior 使用視差先驗(yàn)增強(qiáng)立體圖像的空間分辨率
306.Environment Upgrade Reinforcement Learning for Non-Differentiable Multi-Stage Pipelines 不可分多階段管道的環(huán)境升級(jí)強(qiáng)化學(xué)習(xí)
307.EPINET: A Fully-Convolutional Neural Network Using Epipolar Geometry for Depth From Light Field Images EPINET：一種全卷積神經(jīng)網(wǎng)絡(luò)，使用對(duì)極幾何學(xué)從光場(chǎng)圖像中提取深度
308.Erase or Fill? Deep Joint Recurrent Rain Removal and Reconstruction in Videos 擦除還是填充？視頻中的深層關(guān)節(jié)經(jīng)常性除雨和重建
309.Estimation of Camera Locations in Highly Corrupted Scenarios: All About That Base, No Shape Trouble 高度損壞的場(chǎng)景中攝像機(jī)位置的估計(jì)：關(guān)于該基準(zhǔn)的所有信息，沒有形狀問題
310.Event-Based Vision Meets Deep Learning on Steering Prediction for Self-Driving Cars 基于事件的愿景與無(wú)人駕駛汽車轉(zhuǎn)向預(yù)測(cè)的深度學(xué)習(xí)相遇
311.Every Smile Is Unique: Landmark-Guided Diverse Smile Generation 每個(gè)微笑都是獨(dú)一無(wú)二的：具有地標(biāo)性的多樣化微笑產(chǎn)生
312.Excitation Backprop for RNNs RNN的激勵(lì)反向傳播
313.Explicit Loss-Error-Aware Quantization for Low-Bit Deep Neural Networks 低位深度神經(jīng)網(wǎng)絡(luò)的明確的丟失錯(cuò)誤感知量化
314.Exploiting Transitivity for Learning Person Re-Identification Models on a Budget 在預(yù)算中利用可傳遞性學(xué)習(xí)人的重新識(shí)別模型
315.Exploit the Unknown Gradually: One-Shot Video-Based Person Re-Identification by Stepwise Learning 逐步利用未知：通過(guò)逐步學(xué)習(xí)對(duì)基于視頻的一擊式人員進(jìn)行重新識(shí)別
316.Exploring Disentangled Feature Representation Beyond Face Identification 探索超越人臉識(shí)別的非糾纏特征表示
317.Extreme 3D Face Reconstruction: Seeing Through Occlusions 極端3D面部重建：透視遮擋
318.Eye In-Painting With Exemplar Generative Adversarial Networks 使用示例性生成對(duì)抗網(wǎng)絡(luò)進(jìn)行眼睛內(nèi)畫
319.Face Aging With Identity-Preserved Conditional Generative Adversarial Networks 保留身份的條件生成對(duì)抗網(wǎng)絡(luò)的面孔老化
320.FaceID-GAN: Learning a Symmetry Three-Player GAN for Identity-Preserving Face Synthesis FaceID-GAN：學(xué)習(xí)對(duì)稱三層GAN來(lái)保持身份的人臉合成
321.Facelet-Bank for Fast Portrait Manipulation Facelet-Bank用于快速人像操作
322.Facial Expression Recognition by De-Expression Residue Learning 去表達(dá)殘基學(xué)習(xí)的面部表情識(shí)別
323.Factoring Shape, Pose, and Layout From the 2D Image of a 3D Scene 從3D場(chǎng)景的2D圖像分解形狀，姿勢(shì)和布局
324.Fast and Accurate Online Video Object Segmentation via Tracking Parts 通過(guò)跟蹤部件快速，準(zhǔn)確地在線分割視頻對(duì)象
325.Fast and Accurate Single Image Super-Resolution via Information Distillation Network 通過(guò)信息蒸餾網(wǎng)絡(luò)實(shí)現(xiàn)快速，準(zhǔn)確的單圖像超分辨率
326.Fast and Furious: Real Time End-to-End 3D Detection, Tracking and Motion Forecasting With a Single Convolutional Net 速度與激情：使用單個(gè)卷積網(wǎng)絡(luò)進(jìn)行實(shí)時(shí)端到端3D檢測(cè)，跟蹤和運(yùn)動(dòng)預(yù)測(cè)
327.Fast and Robust Estimation for Unit-Norm Constrained Linear Fitting Problems 單位范數(shù)約束線性擬合問題的快速魯棒估計(jì)
328.Fast End-to-End Trainable Guided Filter 快速的端到端可訓(xùn)練導(dǎo)引濾波器
329.Fast Monte-Carlo Localization on Aerial Vehicles Using Approximate Continuous Belief Representations 使用近似連續(xù)信念表示法對(duì)飛行器進(jìn)行快速蒙特卡洛定位
330.Fast Spectral Ranking for Similarity Search 相似搜索的快速光譜排名
331.Fast Video Object Segmentation by Reference-Guided Mask Propagation 通過(guò)參考引導(dǎo)的遮罩傳播進(jìn)行快速視頻對(duì)象分割
332.FeaStNet: Feature-Steered Graph Convolutions for 3D Shape Analysis FeaStNet：用于3D形狀分析的基于特征的圖卷積
333.Feature Generating Networks for Zero-Shot Learning 零發(fā)學(xué)習(xí)的特征生成網(wǎng)絡(luò)
334.Feature Mapping for Learning Fast and Accurate 3D Pose Inference From Synthetic Images 從合成圖像中快速，準(zhǔn)確地學(xué)習(xí)3D姿勢(shì)推斷的特征映射
335.Feature Quantization for Defending Against Distortion of Images 防止圖像失真的特征量化
336.Feature Selective Networks for Object Detection 用于目標(biāo)檢測(cè)的特征選擇網(wǎng)絡(luò)
337.Features for Multi-Target Multi-Camera Tracking and Re-Identification 多目標(biāo)多攝像機(jī)跟蹤和重新識(shí)別功能
338.Feature Space Transfer for Data Augmentation 特征空間傳輸以增強(qiáng)數(shù)據(jù)
339.Feature Super-Resolution: Make Machine See More Clearly 功能超高分辨率：使機(jī)器更加清晰
340.Feedback-Prop: Convolutional Neural Network Inference Under Partial Evidence 反饋支持：部分證據(jù)下的卷積神經(jīng)網(wǎng)絡(luò)推理
341.Few-Shot Image Recognition by Predicting Parameters From Activations 通過(guò)預(yù)測(cè)激活參數(shù)來(lái)進(jìn)行少量圖像識(shí)別
342.FFNet: Video Fast-Forwarding via Reinforcement Learning FFNet：通過(guò)強(qiáng)化學(xué)習(xí)進(jìn)行視頻快速轉(zhuǎn)發(fā)
343.Fight Ill-Posedness With Ill-Posedness: Single-Shot Variational Depth Super-Resolution From Shading 與病態(tài)對(duì)抗病態(tài)：陰影的單發(fā)變化深度超級(jí)分辨率
344.Finding Beans in Burgers: Deep Semantic-Visual Embedding With Localization 在漢堡中尋找豆子：具有本地化功能的深度語(yǔ)義視覺嵌入
345.Finding “It”: Weakly-Supervised Reference-Aware Visual Grounding in Instructional Videos 找到“它”：教學(xué)視頻中受弱監(jiān)督的參考感知的視覺基礎(chǔ)
346.Finding Tiny Faces in the Wild With Generative Adversarial Network 利用生成對(duì)抗網(wǎng)絡(luò)在野外尋找小臉
347.Fine-Grained Video Captioning for Sports Narrative 體育敘事的細(xì)粒度視頻字幕
348.First-Person Hand Action Benchmark With RGB-D Videos and 3D Hand Pose Annotations 具有RGB-D視頻和3D手勢(shì)注釋的第一人稱手勢(shì)基準(zhǔn)
349.Five-Point Fundamental Matrix Estimation for Uncalibrated Cameras 未校準(zhǔn)相機(jī)的五點(diǎn)基本矩陣估計(jì)
350.FlipDial: A Generative Model for Two-Way Visual Dialogue FlipDial：雙向視覺對(duì)話的生成模型

論文概要

351.Flow Guided Recurrent Neural Encoder for Video Salient Object Detection 流導(dǎo)向的遞歸神經(jīng)編碼器，用于視頻顯著目標(biāo)檢測(cè)
352.Focal Visual-Text Attention for Visual Question Answering 視覺問題解答的焦點(diǎn)視覺文本注意
353.Focus Manipulation Detection via Photometric Histogram Analysis 通過(guò)光度直方圖分析進(jìn)行焦點(diǎn)操縱檢測(cè)
354.FoldingNet: Point Cloud Auto-Encoder via Deep Grid Deformation FoldingNet：通過(guò)深層網(wǎng)格變形的點(diǎn)云自動(dòng)編碼器
355.Fooling Vision and Language Models Despite Localization and Attention Mechanism 盡管存在本地化和注意力機(jī)制，但仍會(huì)愚弄視覺和語(yǔ)言模型
356.FOTS: Fast Oriented Text Spotting With a Unified Network FOTS：使用統(tǒng)一網(wǎng)絡(luò)快速定位文本
357.Frame-Recurrent Video Super-Resolution 幀循環(huán)視頻超分辨率
358.Free Supervision From Video Games 電子游戲免費(fèi)監(jiān)督
359.From Lifestyle Vlogs to Everyday Interactions 從生活時(shí)尚博客到日常互動(dòng)
360.From Source to Target and Back: Symmetric Bi-Directional Adaptive GAN 從源到目標(biāo)再到目標(biāo)：對(duì)稱雙向自適應(yīng)GAN
361.Frustum PointNets for 3D Object Detection From RGB-D Data 用于從RGB-D數(shù)據(jù)進(jìn)行3D對(duì)象檢測(cè)的Frustum PointNets
362.FSRNet: End-to-End Learning Face Super-Resolution With Facial Priors FSRNet：具有面部先驗(yàn)的端到端學(xué)習(xí)面孔超分辨率
363.Fully Convolutional Adaptation Networks for Semantic Segmentation 用于語(yǔ)義分割的全卷積自適應(yīng)網(wǎng)絡(luò)
364.Functional Map of the World 世界功能地圖
365.Fusing Crowd Density Maps and Visual Object Trackers for People Tracking in Crowd Scenes 融合人群密度圖和視覺對(duì)象跟蹤器以在人群場(chǎng)景中進(jìn)行人跟蹤
366.Future Frame Prediction for Anomaly Detection - A New Baseline 異常檢測(cè)的未來(lái)幀預(yù)測(cè)-新基準(zhǔn)
367.Future Person Localization in First-Person Videos 第一人稱視頻中的未來(lái)人本地化
368.GAGAN: Geometry-Aware Generative Adversarial Networks GAGAN：幾何感知生成對(duì)抗網(wǎng)絡(luò)
369.GANerated Hands for Real-Time 3D Hand Tracking From Monocular RGB 用于單眼RGB的實(shí)時(shí)3D手跟蹤的分層手
370.Gated Fusion Network for Single Image Dehazing 門控融合網(wǎng)絡(luò)用于單圖像去霧
371.Gaze Prediction in Dynamic 360deg Immersive Videos 動(dòng)態(tài)360度沉浸式視頻中的注視預(yù)測(cè)
372.Generalized Zero-Shot Learning via Synthesized Examples 通過(guò)綜合實(shí)例進(jìn)行廣義零槍學(xué)習(xí)
373.Generate to Adapt: Aligning Domains Using Generative Adversarial Networks 生成以適應(yīng)：使用生成對(duì)抗網(wǎng)絡(luò)調(diào)整域
374.Generating a Fusion Image: One’s Identity and Another’s Shape 生成融合圖像：一個(gè)人的身份和另一個(gè)人的形狀
375.Generating Synthetic X-Ray Images of a Person From the Surface Geometry 從表面幾何形狀生成人的合成X射線圖像
376.Generative Adversarial Image Synthesis With Decision Tree Latent Controller 決策樹潛在控制器的對(duì)抗式生成圖像綜合
377.Generative Adversarial Learning Towards Fast Weakly Supervised Detection 生成對(duì)抗性學(xué)習(xí)，實(shí)現(xiàn)快速弱監(jiān)督檢測(cè)
378.Generative Adversarial Perturbations 生成對(duì)抗性擾動(dòng)
379.Generative Image Inpainting With Contextual Attention 具有上下文注意的生成圖像修復(fù)
380.Generative Modeling Using the Sliced Wasserstein Distance 使用切片Wasserstein距離進(jìn)行生成建模
381.Geometric Multi-Model Fitting With a Convex Relaxation Algorithm 凸松弛算法進(jìn)行幾何多模型擬合
382.Geometric Robustness of Deep Networks: Analysis and Improvement 深度網(wǎng)絡(luò)的幾何魯棒性：分析和改進(jìn)
383.Geometry Aware Constrained Optimization Techniques for Deep Learning 深度學(xué)習(xí)的幾何感知約束優(yōu)化技術(shù)
384.Geometry-Aware Deep Network for Single-Image Novel View Synthesis 用于單圖像新穎視圖合成的幾何感知深度網(wǎng)絡(luò)
385.Geometry-Aware Learning of Maps for Camera Localization 用于相機(jī)定位的地圖的幾何感知學(xué)習(xí)
386.Geometry-Aware Network for Non-Rigid Shape Prediction From a Single View 從單個(gè)視圖進(jìn)行非剛性形狀預(yù)測(cè)的幾何感知網(wǎng)絡(luò)
387.Geometry-Aware Scene Text Detection With Instance Transformation Network 具有實(shí)例轉(zhuǎn)換網(wǎng)絡(luò)的幾何感知場(chǎng)景文本檢測(cè)
388.Geometry Guided Convolutional Neural Networks for Self-Supervised Video Representation Learning 幾何指導(dǎo)的卷積神經(jīng)網(wǎng)絡(luò)用于自指導(dǎo)視頻表示學(xué)習(xí)
389.GeoNet: Geometric Neural Network for Joint Depth and Surface Normal Estimation GeoNet：用于聯(lián)合深度和表面法線估計(jì)的幾何神經(jīng)網(wǎng)絡(luò)
390.GeoNet: Unsupervised Learning of Dense Depth, Optical Flow and Camera Pose GeoNet：密集深度，光流和相機(jī)姿勢(shì)的無(wú)監(jiān)督學(xué)習(xí)
391.Gesture Recognition: Focus on the Hands 手勢(shì)識(shí)別：專注于手
392.Gibson Env: Real-World Perception for Embodied Agents 吉布森環(huán)境（Gibson Env）：現(xiàn)實(shí)世界對(duì)特工的看法
393.Glimpse Clouds: Human Activity Recognition From Unstructured Feature Points 瞥見云：來(lái)自非結(jié)構(gòu)化特征點(diǎn)的人類活動(dòng)識(shí)別
394.Globally Optimal Inlier Set Maximization for Atlanta Frame Estimation 亞特蘭大幀估計(jì)的全局最優(yōu)Inlier集最大化
395.Global Versus Localized Generative Adversarial Nets 全球與本地化生成對(duì)抗網(wǎng)
396.Going From Image to Video Saliency: Augmenting Image Salience With Dynamic Attentional Push 從圖像到視頻顯著性：通過(guò)動(dòng)態(tài)注意力推送來(lái)增強(qiáng)圖像顯著性
397.Good View Hunting: Learning Photo Composition From Dense View Pairs 良好的視野狩獵：從密集的視野對(duì)中學(xué)習(xí)照片構(gòu)圖
398.GraphBit: Bitwise Interaction Mining via Deep Reinforcement Learning GraphBit：通過(guò)深度強(qiáng)化學(xué)習(xí)進(jìn)行按位交互挖掘
399.Graph-Cut RANSAC 圖切RANSAC
400.Grounding Referring Expressions in Images by Variational Context 通過(guò)變體上下文使圖像中的指稱表達(dá)接地

論文概要

401.GroupCap: Group-Based Image Captioning With Structured Relevance and Diversity Constraints GroupCap：具有結(jié)構(gòu)相關(guān)性和多樣性約束的基于組的圖像字幕
402.Group Consistent Similarity Learning via Deep CRF for Person Re-Identification 通過(guò)深度CRF進(jìn)行群體一致性相似性學(xué)習(xí)以進(jìn)行人員重新識(shí)別
403.Guided Proofreading of Automatic Segmentations for Connectomics Connectomics自動(dòng)細(xì)分的指導(dǎo)性校對(duì)
404.Guide Me: Interacting With Deep Networks 指導(dǎo)我：與深度網(wǎng)絡(luò)互動(dòng)
405.GVCNN: Group-View Convolutional Neural Networks for 3D Shape Recognition GVCNN：用于3D形狀識(shí)別的組視圖卷積神經(jīng)網(wǎng)絡(luò)
406.Hallucinated-IQA: No-Reference Image Quality Assessment via Adversarial Learning 幻覺IQA：通過(guò)對(duì)抗學(xué)習(xí)進(jìn)行無(wú)參考圖像質(zhì)量評(píng)估
407.Hand PointNet: 3D Hand Pose Estimation Using Point Sets Hand PointNet：使用點(diǎn)集的3D手姿估計(jì)
408.Harmonious Attention Network for Person Re-Identification 重新識(shí)別人的和諧注意網(wǎng)絡(luò)
409.HashGAN: Deep Learning to Hash With Pair Conditional Wasserstein GAN HashGAN：深度學(xué)習(xí)與有條件的Wasserstein GAN配對(duì)
410.Hashing as Tie-Aware Learning to Rank 散列為領(lǐng)帶感知學(xué)習(xí)排名
411.HATS: Histograms of Averaged Time Surfaces for Robust Event-Based Object Classification HATS：魯棒的基于事件的對(duì)象分類的平均時(shí)間表面直方圖
412.Hierarchical Novelty Detection for Visual Object Recognition 視覺對(duì)象識(shí)別的層次新穎性檢測(cè)
413.Hierarchical Recurrent Attention Networks for Structured Online Maps 結(jié)構(gòu)化在線地圖的分層遞歸注意網(wǎng)絡(luò)
414.High-Order Tensor Regularization With Application to Attribute Ranking 高階張量正則化及其在屬性排序中的應(yīng)用
415.High Performance Visual Tracking With Siamese Region Proposal Network 連體區(qū)域提案網(wǎng)絡(luò)的高性能視覺跟蹤
416.High-Resolution Image Synthesis and Semantic Manipulation With Conditional GANs 有條件GAN的高分辨率圖像合成和語(yǔ)義處理
417.High-Speed Tracking With Multi-Kernel Correlation Filters 利用多核相關(guān)濾波器進(jìn)行高速跟蹤
418.HSA-RNN: Hierarchical Structure-Adaptive RNN for Video Summarization HSA-RNN：用于視頻匯總的分層結(jié)構(gòu)自適應(yīng)RNN
419.Human Appearance Transfer 人的外觀轉(zhuǎn)移
420.Human-Centric Indoor Scene Synthesis Using Stochastic Grammar 基于隨機(jī)語(yǔ)法的以人為中心的室內(nèi)場(chǎng)景合成
421.Human Pose Estimation With Parsing Induced Learner 解析誘導(dǎo)學(xué)習(xí)者的人體姿勢(shì)估計(jì)
422.Human Semantic Parsing for Person Re-Identification 用于人員重新識(shí)別的人類語(yǔ)義解析
423.Hybrid Camera Pose Estimation 混合相機(jī)姿勢(shì)估計(jì)
424.HydraNets: Specialized Dynamic Architectures for Efficient Inference HydraNets：高效推理的專用動(dòng)態(tài)架構(gòu)
425.Hyperparameter Optimization for Tracking With Continuous Deep Q-Learning 連續(xù)深度Q學(xué)習(xí)的超參數(shù)優(yōu)化跟蹤
426.ICE-BA: Incremental, Consistent and Efficient Bundle Adjustment for Visual-Inertial SLAM ICE-BA：視覺慣性SLAM的增量，一致和高效的捆綁包調(diào)整
427.Illuminant Spectra-Based Source Separation Using Flash Photography 使用閃光燈攝影的基于光譜的光源分離
428.Im2Flow: Motion Hallucination From Static Images for Action Recognition Im2Flow：從靜態(tài)圖像進(jìn)行動(dòng)作幻覺以進(jìn)行動(dòng)作識(shí)別
429.Im2Pano3D: Extrapolating 360deg Structure and Semantics Beyond the Field of View Im2Pano3D：超越視野，外推360度結(jié)構(gòu)和語(yǔ)義
430.Im2Struct: Recovering 3D Shape Structure From a Single RGB Image Im2Struct：從單個(gè)RGB圖像中恢復(fù)3D形狀結(jié)構(gòu)
431.Image Blind Denoising With Generative Adversarial Network Based Noise Modeling 基于生成對(duì)抗網(wǎng)絡(luò)的噪聲建模的圖像盲去噪
432.Image Collection Pop-Up: 3D Reconstruction and Clustering of Rigid and Non-Rigid Categories 圖像集合彈出窗口：剛性和非剛性類別的3D重構(gòu)和聚類
433.Image Correction via Deep Reciprocating HDR Transformation 通過(guò)深度往復(fù)HDR變換進(jìn)行圖像校正
434.Image Generation From Scene Graphs 從場(chǎng)景圖生成圖像
435.Image-Image Domain Adaptation With Preserved Self-Similarity and Domain-Dissimilarity for Person Re-Identification 保留人的自我相似性和域相似性的圖像-圖像域自適應(yīng)
436.Image Restoration by Estimating Frequency Distribution of Local Patches 通過(guò)估計(jì)局部補(bǔ)丁的頻率分布來(lái)恢復(fù)圖像
437.Image Super-Resolution via Dual-State Recurrent Networks 通過(guò)雙狀態(tài)循環(huán)網(wǎng)絡(luò)實(shí)現(xiàn)圖像超分辨率
438.Image to Image Translation for Domain Adaptation 圖像到圖像翻譯以進(jìn)行域自適應(yīng)
439.Importance Weighted Adversarial Nets for Partial Domain Adaptation 局部域自適應(yīng)的重要性加權(quán)對(duì)抗網(wǎng)
440.Improved Fusion of Visual and Language Representations by Dense Symmetric Co-Attention for Visual Question Answering 密集的對(duì)稱共同注意對(duì)視覺問題的回答，改善了視覺和語(yǔ)言表示的融合
441.Improved Lossy Image Compression With Priming and Spatially Adaptive Bit Rates for Recurrent Networks 面向遞歸網(wǎng)絡(luò)的具有啟動(dòng)和空間自適應(yīng)位速率的改進(jìn)的有損圖像壓縮
442.Improvements to Context Based Self-Supervised Learning 基于上下文的自我監(jiān)督學(xué)習(xí)的改進(jìn)
443.Improving Color Reproduction Accuracy on Cameras 提高相機(jī)的色彩還原精度
444.Improving Landmark Localization With Semi-Supervised Learning 通過(guò)半監(jiān)督學(xué)習(xí)改善地標(biāo)本地化
445.Improving Object Localization With Fitness NMS and Bounded IoU Loss 使用Fitness NMS和有限的IoU損失改善對(duì)象定位
446.Improving Occlusion and Hard Negative Handling for Single-Stage Pedestrian Detectors 改善單階段行人探測(cè)器的遮擋和硬負(fù)處理
447.Independently Recurrent Neural Network (IndRNN): Building a Longer and Deeper RNN 獨(dú)立循環(huán)神經(jīng)網(wǎng)絡(luò)（IndRNN）：構(gòu)建更長(zhǎng)更深的RNN
448.Indoor RGB-D Compass From a Single Line and Plane 單線和平面的室內(nèi)RGB-D指南針
449.Inference in Higher Order MRF-MAP Problems With Small and Large Cliques 帶有小集團(tuán)的高階MRF-MAP問題的推論
450.Inferring Light Fields From Shadows 從陰影推斷光場(chǎng)

論文概要

451.Inferring Semantic Layout for Hierarchical Text-to-Image Synthesis 推理語(yǔ)義布局以實(shí)現(xiàn)文本到圖像的分層合成
452.Inferring Shared Attention in Social Scene Videos 推斷社交場(chǎng)景視頻中的共享注意力
453.InLoc: Indoor Visual Localization With Dense Matching and View Synthesis InLoc：具有密集匹配和視圖綜合的室內(nèi)視覺本地化
454.In-Place Activated BatchNorm for Memory-Optimized Training of DNNs 就地激活的BatchNorm，用于DNN的內(nèi)存優(yōu)化訓(xùn)練
455.Instance Embedding Transfer to Unsupervised Video Object Segmentation 實(shí)例嵌入轉(zhuǎn)移到無(wú)監(jiān)督視頻對(duì)象分割
456.Interactive Image Segmentation With Latent Diversity 具有潛在多樣性的交互式圖像分割
457.Interleaved Structured Sparse Convolutional Neural Networks 交錯(cuò)結(jié)構(gòu)的稀疏卷積神經(jīng)網(wǎng)絡(luò)
458.Interpretable Convolutional Neural Networks 可解釋的卷積神經(jīng)網(wǎng)絡(luò)
459.Interpretable Video Captioning via Trajectory Structured Localization 通過(guò)軌跡結(jié)構(gòu)化本地化可解釋的視頻字幕
460.Interpret Neural Networks by Identifying Critical Data Routing Paths 通過(guò)識(shí)別關(guān)鍵數(shù)據(jù)路由路徑來(lái)解釋神經(jīng)網(wǎng)絡(luò)
461.Intrinsic Image Transformation via Scale Space Decomposition 通過(guò)尺度空間分解的本征圖像變換
462.Inverse Composition Discriminative Optimization for Point Cloud Registration 點(diǎn)云配準(zhǔn)的逆組合判別優(yōu)化
463.InverseFaceNet: Deep Monocular Inverse Face Rendering InverseFaceNet：深單目反面渲染
464.IQA: Visual Question Answering in Interactive Environments IQA：交互環(huán)境中的視覺問題解答
465.ISTA-Net: Interpretable Optimization-Inspired Deep Network for Image Compressive Sensing ISTA-Net：基于可解釋性優(yōu)化的深度網(wǎng)絡(luò)，用于圖像壓縮傳感
466.Iterative Learning With Open-Set Noisy Labels 開放式嘈雜標(biāo)簽的迭代學(xué)習(xí)
467.Iterative Visual Reasoning Beyond Convolutions 超越卷積的迭代視覺推理
468.IVQA: Inverse Visual Question Answering IVQA：逆向視覺問答
469.Jerk-Aware Video Acceleration Magnification 挺舉感知視頻加速倍率
470.Joint Cuts and Matching of Partitions in One Graph 一幅圖中的聯(lián)合切割和分區(qū)匹配
471.Jointly Localizing and Describing Events for Dense Video Captioning 聯(lián)合本地化和描述用于密集視頻字幕的事件
472.Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation 聯(lián)合優(yōu)化數(shù)據(jù)增強(qiáng)和網(wǎng)絡(luò)培訓(xùn)：人體姿勢(shì)估計(jì)中的對(duì)抗性數(shù)據(jù)增強(qiáng)
473.Joint Optimization Framework for Learning With Noisy Labels 帶有噪音標(biāo)簽的聯(lián)合優(yōu)化學(xué)習(xí)框架
474.Joint Pose and Expression Modeling for Facial Expression Recognition 面部表情識(shí)別的聯(lián)合姿勢(shì)和表情建模
475.Kernelized Subspace Pooling for Deep Local Descriptors 深度本地描述符的內(nèi)核化子空間池
476.KIPPI: KInetic Polygonal Partitioning of Images KIPPI：圖像的運(yùn)動(dòng)多邊形分割
477.Knowledge Aided Consistency for Weakly Supervised Phrase Grounding 弱監(jiān)督短語(yǔ)接地的知識(shí)輔助一致性
478.Label Denoising Adversarial Network (LDAN) for Inverse Lighting of Faces 用于面部反照的標(biāo)簽去噪對(duì)抗網(wǎng)絡(luò)（LDAN）
479.LAMV: Learning to Align and Match Videos With Kernelized Temporal Layers LAMV：學(xué)習(xí)將視頻與內(nèi)核時(shí)間層對(duì)齊和匹配
480.Language-Based Image Editing With Recurrent Attentive Models 基于循環(huán)注意力模型的基于語(yǔ)言的圖像編輯
481.Large-Scale Distance Metric Learning With Uncertainty 不確定性的大規(guī)模遠(yuǎn)程度量學(xué)習(xí)
482.Large Scale Fine-Grained Categorization and Domain-Specific Transfer Learning 大規(guī)模細(xì)粒度分類和特定領(lǐng)域轉(zhuǎn)移學(xué)習(xí)
483.Large-Scale Point Cloud Semantic Segmentation With Superpoint Graphs 超點(diǎn)圖的大規(guī)模點(diǎn)云語(yǔ)義分割
484.Latent RANSAC 潛在的RANSAC
485.LayoutNet: Reconstructing the 3D Room Layout From a Single RGB Image LayoutNet：從單個(gè)RGB圖像重建3D房間布局
486.LDMNet: Low Dimensional Manifold Regularized Neural Networks LDMNet：低維流形正則化神經(jīng)網(wǎng)絡(luò)
487.Lean Multiclass Crowdsourcing 精益多類眾包
488.Learned Shape-Tailored Descriptors for Segmentation 習(xí)得的形狀定制描述符用于細(xì)分
489.Learning 3D Shape Completion From Laser Scan Data With Weak Supervision 通過(guò)弱監(jiān)督從激光掃描數(shù)據(jù)中學(xué)習(xí)3D形狀完成
490.Learning a Complete Image Indexing Pipeline 學(xué)習(xí)完整的圖像索引管道
491.Learning a Discriminative Feature Network for Semantic Segmentation 學(xué)習(xí)用于語(yǔ)義分割的判別特征網(wǎng)絡(luò)
492.Learning a Discriminative Filter Bank Within a CNN for Fine-Grained Recognition 學(xué)習(xí)CNN中的判別式濾波器組以進(jìn)行細(xì)粒度識(shí)別
493.Learning a Discriminative Prior for Blind Image Deblurring 學(xué)習(xí)判別先驗(yàn)盲圖像去模糊
494.Learning and Using the Arrow of Time 學(xué)習(xí)和使用時(shí)間之箭
495.Learning Answer Embeddings for Visual Question Answering 學(xué)習(xí)視覺視覺答案的答案嵌入
496.Learning a Single Convolutional Super-Resolution Network for Multiple Degradations 學(xué)習(xí)單個(gè)卷積超分辨率網(wǎng)絡(luò)以進(jìn)行多次降級(jí)
497.Learning Attentions: Residual Attentional Siamese Network for High Performance Online Visual Tracking 學(xué)習(xí)注意力：殘余注意力連體網(wǎng)絡(luò)，用于高性能在線視覺跟蹤
498.Learning Attribute Representations With Localization for Flexible Fashion Search 通過(guò)本地化學(xué)習(xí)屬性表示以實(shí)現(xiàn)靈活的時(shí)尚搜索
499.Learning by Asking Questions 提問學(xué)習(xí)
500.Learning Compact Recurrent Neural Networks With Block-Term Tensor Decomposition 通過(guò)塊期張量分解學(xué)習(xí)緊湊型遞歸神經(jīng)網(wǎng)絡(luò)

論文概要

501.Learning Compositional Visual Concepts With Mutual Consistency 相互一致地學(xué)習(xí)構(gòu)圖視覺概念
502.Learning Compressible 360deg Video Isomers 學(xué)習(xí)可壓縮的360deg視頻異構(gòu)體
503.“Learning-Compression” Algorithms for Neural Net Pruning 神經(jīng)網(wǎng)絡(luò)修剪的“學(xué)習(xí)-壓縮”算法
504.Learning Convolutional Networks for Content-Weighted Image Compression 學(xué)習(xí)卷積網(wǎng)絡(luò)進(jìn)行內(nèi)容加權(quán)圖像壓縮
505.Learning Deep Descriptors With Scale-Aware Triplet Networks 使用可感知規(guī)模的三重態(tài)網(wǎng)絡(luò)學(xué)習(xí)深度描述符
506.Learning Deep Models for Face Anti-Spoofing: Binary or Auxiliary Supervision 學(xué)習(xí)深度模型進(jìn)行面部反欺騙：二進(jìn)制或輔助監(jiān)督
507.Learning Deep Sketch Abstraction 學(xué)習(xí)深度素描抽象
508.Learning Deep Structured Active Contours End-to-End 端到端學(xué)習(xí)深度結(jié)構(gòu)化活動(dòng)輪廓
509.Learning Depth From Monocular Videos Using Direct Methods 使用直接方法從單眼視頻中學(xué)習(xí)深度
510.Learning Descriptor Networks for 3D Shape Synthesis and Analysis 用于3D形狀合成和分析的學(xué)習(xí)描述符網(wǎng)絡(luò)
511.Learning Distributions of Shape Trajectories From Longitudinal Datasets: A Hierarchical Model on a Manifold of Diffeomorphisms 從縱向數(shù)據(jù)集學(xué)習(xí)形狀軌跡的分布：Diffeomorphisms流形的層次模型。
512.Learning Dual Convolutional Neural Networks for Low-Level Vision 學(xué)習(xí)雙卷積神經(jīng)網(wǎng)絡(luò)以實(shí)現(xiàn)低視力
513.Learning Face Age Progression: A Pyramid Architecture of GANs 學(xué)習(xí)面部年齡發(fā)展：GAN的金字塔體系結(jié)構(gòu)
514.Learning Facial Action Units From Web Images With Scalable Weakly Supervised Clustering 通過(guò)可擴(kuò)展的弱監(jiān)督聚類從Web圖像中學(xué)習(xí)面部動(dòng)作單元
515.Learning for Disparity Estimation Through Feature Constancy 通過(guò)特征恒定學(xué)習(xí)差異估計(jì)
516.Learning From Millions of 3D Scans for Large-Scale 3D Face Recognition 從數(shù)百萬(wàn)的3D掃描中學(xué)習(xí)以進(jìn)行大規(guī)模3D人臉識(shí)別
517.Learning From Noisy Web Data With Category-Level Supervision 通過(guò)類別級(jí)監(jiān)督從嘈雜的Web數(shù)據(jù)中學(xué)習(xí)
518.Learning From Synthetic Data: Addressing Domain Shift for Semantic Segmentation 從合成數(shù)據(jù)中學(xué)習(xí)：解決語(yǔ)義語(yǔ)義分割的域移位
519.Learning Generative ConvNets via Multi-Grid Modeling and Sampling 通過(guò)多網(wǎng)格建模和采樣學(xué)習(xí)生成式ConvNet
520.Learning Globally Optimized Object Detector via Policy Gradient 通過(guò)策略梯度學(xué)習(xí)全局優(yōu)化的對(duì)象檢測(cè)器
521.Learning Intelligent Dialogs for Bounding Box Annotation 學(xué)習(xí)邊界框注釋的智能對(duì)話框
522.Learning Intrinsic Image Decomposition From Watching the World 從觀看世界中學(xué)習(xí)內(nèi)在的圖像分解
523.Learning Latent Super-Events to Detect Multiple Activities in Videos 學(xué)習(xí)潛在的超級(jí)事件以檢測(cè)視頻中的多個(gè)活動(dòng)
524.Learning Less Is More - 6D Camera Localization via 3D Surface Regression 學(xué)會(huì)少即是多-通過(guò)3D表面回歸實(shí)現(xiàn)6D相機(jī)本地化
525.Learning Markov Clustering Networks for Scene Text Detection 學(xué)習(xí)用于場(chǎng)景文本檢測(cè)的馬爾可夫聚類網(wǎng)絡(luò)
526.Learning Monocular 3D Human Pose Estimation From Multi-View Images 從多視圖圖像中學(xué)習(xí)單眼3D人類姿勢(shì)估計(jì)
527.Learning Multi-Instance Enriched Image Representations via Non-Greedy Ratio Maximization of the l1-Norm Distances 通過(guò)l1-Norm距離的非貪心比最大化學(xué)習(xí)多實(shí)例富集圖像表示
528.Learning Patch Reconstructability for Accelerating Multi-View Stereo 學(xué)習(xí)補(bǔ)丁可重構(gòu)性，以加速多視圖立體聲
529.Learning Pixel-Level Semantic Affinity With Image-Level Supervision for Weakly Supervised Semantic Segmentation 通過(guò)圖像級(jí)監(jiān)督學(xué)習(xí)像素級(jí)語(yǔ)義親和度以實(shí)現(xiàn)弱監(jiān)督語(yǔ)義分割
530.Learning Pose Specific Representations by Predicting Different Views 通過(guò)預(yù)測(cè)不同的觀點(diǎn)來(lái)學(xué)習(xí)姿勢(shì)特定表示
531.Learning Rich Features for Image Manipulation Detection 學(xué)習(xí)豐富的圖像操縱檢測(cè)功能
532.Learning Semantic Concepts and Order for Image and Sentence Matching 學(xué)習(xí)語(yǔ)義概念和圖像和句子匹配的順序
533.Learning Spatial-Aware Regressions for Visual Tracking 學(xué)習(xí)用于視覺跟蹤的空間感知回歸
534.Learning Spatial-Temporal Regularized Correlation Filters for Visual Tracking 學(xué)習(xí)時(shí)空正則化相關(guān)濾波器以進(jìn)行視覺跟蹤
535.Learning Steerable Filters for Rotation Equivariant CNNs 學(xué)習(xí)旋轉(zhuǎn)等變CNN的可控濾波器
536.Learning Strict Identity Mappings in Deep Residual Networks 在深度殘差網(wǎng)絡(luò)中學(xué)習(xí)嚴(yán)格的身份映射
537.Learning Structure and Strength of CNN Filters for Small Sample Size Training 小樣本量訓(xùn)練的CNN濾波器的學(xué)習(xí)結(jié)構(gòu)和強(qiáng)度
538.Learning Superpixels With Segmentation-Aware Affinity Loss 通過(guò)分段感知的親和力損失學(xué)習(xí)超像素
539.Learning Time_Memory-Efficient Deep Architectures With Budgeted Super Networks 通過(guò)預(yù)算的超級(jí)網(wǎng)絡(luò)學(xué)習(xí)時(shí)間_內(nèi)存高效的深度架構(gòu)
540.Learning to Act Properly: Predicting and Explaining Affordances From Images 學(xué)習(xí)正確采取行動(dòng)：預(yù)測(cè)和解釋圖像中的負(fù)擔(dān)
541.Learning to Adapt Structured Output Space for Semantic Segmentation 學(xué)習(xí)適應(yīng)結(jié)構(gòu)化輸出空間進(jìn)行語(yǔ)義分割
542.Learning to Compare: Relation Network for Few-Shot Learning 學(xué)習(xí)比較：很少學(xué)習(xí)的關(guān)系網(wǎng)絡(luò)
543.Learning to Detect Features in Texture Images 學(xué)習(xí)檢測(cè)紋理圖像中的特征
544.Learning to Estimate 3D Human Pose and Shape From a Single Color Image 學(xué)習(xí)從單色圖像估計(jì)3D人類姿勢(shì)和形狀
545.Learning to Evaluate Image Captioning 學(xué)習(xí)評(píng)估圖像字幕
546.Learning to Extract a Video Sequence From a Single Motion-Blurred Image 學(xué)習(xí)從單個(gè)運(yùn)動(dòng)模糊圖像中提取視頻序列
547.Learning to Find Good Correspondences 學(xué)習(xí)尋找良好的對(duì)應(yīng)關(guān)系
548.Learning to Generate Time-Lapse Videos Using Multi-Stage Dynamic Generative Adversarial Networks 學(xué)習(xí)使用多階段動(dòng)態(tài)生成對(duì)抗網(wǎng)絡(luò)生成延時(shí)視頻
549.Learning to Localize Sound Source in Visual Scenes 學(xué)習(xí)在視覺場(chǎng)景中定位聲源
550.Learning to Look Around: Intelligently Exploring Unseen Environments for Unknown Tasks 學(xué)習(xí)環(huán)顧四周：智能探索未知任務(wù)未知的環(huán)境

論文概要

551.Learning to Parse Wireframes in Images of Man-Made Environments 學(xué)習(xí)解析人造環(huán)境圖像中的線框
552.Learning to Promote Saliency Detectors 學(xué)習(xí)促進(jìn)顯著性檢測(cè)器
553.Learning to See in the Dark 學(xué)習(xí)在黑暗中看
554.Learning to Segment Every Thing 學(xué)會(huì)分割一切
555.Learning to Sketch With Shortcut Cycle Consistency 學(xué)習(xí)以快捷的周期一致性進(jìn)行素描
556.Learning to Understand Image Blur 學(xué)習(xí)理解圖像模糊
557.Learning Transferable Architectures for Scalable Image Recognition 學(xué)習(xí)可擴(kuò)展的體系結(jié)構(gòu)以實(shí)現(xiàn)可擴(kuò)展的圖像識(shí)別
558.Learning Visual Knowledge Memory Networks for Visual Question Answering 學(xué)習(xí)視覺知識(shí)記憶網(wǎng)絡(luò)以進(jìn)行視覺問答
559.Left-Right Comparative Recurrent Model for Stereo Matching 立體聲匹配的左右比較遞歸模型
560.LEGO: Learning Edge With Geometry All at Once by Watching Videos 樂高：通過(guò)觀看視頻一次學(xué)習(xí)幾何的優(yōu)勢(shì)
561.Leveraging Unlabeled Data for Crowd Counting by Learning to Rank 通過(guò)學(xué)習(xí)排名利用未標(biāo)記的數(shù)據(jù)進(jìn)行人群計(jì)數(shù)
562.LiDAR-Video Driving Dataset: Learning Driving Policies Effectively LiDAR視頻駕駛數(shù)據(jù)集：有效學(xué)習(xí)駕駛策略
563.Light Field Intrinsics With a Deep Encoder-Decoder Network 具有深層編碼器-解碼器網(wǎng)絡(luò)的光場(chǎng)本征
564.Lightweight Probabilistic Deep Networks 輕型概率深度網(wǎng)絡(luò)
565.LIME: Live Intrinsic Material Estimation LIME：實(shí)時(shí)內(nèi)在材料估計(jì)
566.Link and Code: Fast Indexing With Graphs and Compact Regression Codes 鏈接和代碼：使用圖形和緊湊回歸代碼進(jìn)行快速索引
567.Lions and Tigers and Bears: Capturing Non-Rigid, 3D, Articulated Shape From Images 獅子和老虎與熊：從圖像中捕獲非剛性，3D，關(guān)節(jié)形狀
568.LiteFlowNet: A Lightweight Convolutional Neural Network for Optical Flow Estimation LiteFlowNet：用于光流估計(jì)的輕量級(jí)卷積神經(jīng)網(wǎng)絡(luò)
569.Local and Global Optimization Techniques in Graph-Based Clustering 基于圖的聚類中的局部和全局優(yōu)化技術(shù)
570.Local Descriptors Optimized for Average Precision 優(yōu)化本地描述符以實(shí)現(xiàn)平均精度
571.Logo Synthesis and Manipulation With Clustered Generative Adversarial Networks 聚類生成對(duì)抗網(wǎng)絡(luò)的徽標(biāo)合成和操縱
572.Long-Term On-Board Prediction of People in Traffic Scenes Under Uncertainty 不確定情況下交通場(chǎng)景中人員的長(zhǎng)期車載預(yù)測(cè)
573.Look at Boundary: A Boundary-Aware Face Alignment Algorithm 看邊界：邊界感知的人臉對(duì)齊算法
574.Look, Imagine and Match: Improving Textual-Visual Cross-Modal Retrieval With Generative Models 外觀，想象和匹配：使用生成模型改進(jìn)文本視覺跨模態(tài)檢索
575.Lose the Views: Limited Angle CT Reconstruction via Implicit Sinogram Completion 失去視野：通過(guò)隱式Singram完成有限角度CT重建
576.Low-Latency Video Semantic Segmentation 低延遲視頻語(yǔ)義分割
577.Low-Shot Learning From Imaginary Data 虛幻數(shù)據(jù)的低速學(xué)習(xí)
578.Low-Shot Learning With Imprinted Weights 帶有權(quán)重的低速學(xué)習(xí)
579.Low-Shot Learning With Large-Scale Diffusion 大規(guī)模擴(kuò)散的低射學(xué)習(xí)
580.LSTM Pose Machines LSTM姿勢(shì)機(jī)
581.M3: Multimodal Memory Modelling for Video Captioning M3：用于視頻字幕的多模式內(nèi)存建模
582.Making Convolutional Networks Recurrent for Visual Sequence Learning 使卷積網(wǎng)絡(luò)循環(huán)進(jìn)行視覺序列學(xué)習(xí)
583.Manifold Learning in Quotient Spaces 商空間中的流形學(xué)習(xí)
584.MapNet: An Allocentric Spatial Memory for Mapping Environments MapNet：映射環(huán)境的同心圓空間內(nèi)存
585.Mask-Guided Contrastive Attention Model for Person Re-Identification 面罩引導(dǎo)的對(duì)比注意模型用于人員重新識(shí)別
586.MaskLab: Instance Segmentation by Refining Object Detection With Semantic and Direction Features MaskLab：通過(guò)語(yǔ)義和方向特征完善對(duì)象檢測(cè)來(lái)實(shí)現(xiàn)實(shí)例分割
587.Matching Adversarial Networks 匹配的對(duì)抗網(wǎng)絡(luò)
588.Matching Pixels Using Co-Occurrence Statistics 使用共現(xiàn)統(tǒng)計(jì)匹配像素
589.Matryoshka Networks: Predicting 3D Geometry via Nested Shape Layers Matryoshka Networks：通過(guò)嵌套形狀層預(yù)測(cè)3D幾何
590.MAttNet: Modular Attention Network for Referring Expression Comprehension MAttNet：用于引用表達(dá)理解的模塊化注意網(wǎng)絡(luò)
591.Maximum Classifier Discrepancy for Unsupervised Domain Adaptation 無(wú)監(jiān)督域自適應(yīng)的最大分類器差異
592.Mean-Variance Loss for Deep Age Estimation From a Face 從臉部進(jìn)行深度估計(jì)的均值方差損失
593.MegaDepth: Learning Single-View Depth Prediction From Internet Photos MegaDepth：從互聯(lián)網(wǎng)照片中學(xué)習(xí)單視圖深度預(yù)測(cè)
594.MegDet: A Large Mini-Batch Object Detector MegDet：大型小批量物體檢測(cè)器
595.Memory Based Online Learning of Deep Representations From Video Streams 基于內(nèi)存的視頻流深度表示在線學(xué)習(xí)
596.Memory Matching Networks for One-Shot Image Recognition 一鍵式圖像識(shí)別的內(nèi)存匹配網(wǎng)絡(luò)
597.Mesoscopic Facial Geometry Inference Using Deep Neural Networks 使用深層神經(jīng)網(wǎng)絡(luò)的介觀面部幾何推理
598.MiCT: Mixed 3D_2D Convolutional Tube for Human Action Recognition MiCT：用于人類動(dòng)作識(shí)別的混合3D_2D卷積管
599.Min-Entropy Latent Model for Weakly Supervised Object Detection 弱熵目標(biāo)檢測(cè)的最小熵潛模型
600.Mining on Manifolds: Metric Learning Without Labels 流形上的挖掘：無(wú)標(biāo)簽的公制學(xué)習(xí)

論文概要

601.Mining Point Cloud Local Structures by Kernel Correlation and Graph Pooling 基于核相關(guān)和圖池的挖掘點(diǎn)云局部結(jié)構(gòu)
602.Missing Slice Recovery for Tensors Using a Low-Rank Model in Embedded Space 在嵌入式空間中使用低秩模型進(jìn)行張量缺失切片恢復(fù)
603.Mix and Match Networks: Encoder-Decoder Alignment for Zero-Pair Image Translation 混合和匹配網(wǎng)絡(luò)：零對(duì)圖像轉(zhuǎn)換的編碼器-解碼器對(duì)準(zhǔn)
604.MobileNetV2: Inverted Residuals and Linear Bottlenecks MobileNetV2：殘差和線性瓶頸
605.Mobile Video Object Detection With Temporally-Aware Feature Maps 具有臨時(shí)感知特征圖的移動(dòng)視頻對(duì)象檢測(cè)
606.MoCoGAN: Decomposing Motion and Content for Video Generation MoCoGAN：分解運(yùn)動(dòng)和內(nèi)容以生成視頻
607.Modeling Facial Geometry Using Compositional VAEs 使用合成VAE對(duì)面部幾何建模
608.Modifying Non-Local Variations Across Multiple Views 跨多個(gè)視圖修改非局部變化
609.Modulated Convolutional Networks 調(diào)制卷積網(wǎng)絡(luò)
610.MoNet: Deep Motion Exploitation for Video Object Segmentation MoNet：用于視頻對(duì)象分割的深度運(yùn)動(dòng)開發(fā)
611.MoNet: Moments Embedding Network MoNet：時(shí)刻嵌入網(wǎng)絡(luò)
612.Monocular 3D Pose and Shape Estimation of Multiple People in Natural Scenes - The Importance of Multiple Scene Constraints 自然場(chǎng)景中多人的單眼3D姿勢(shì)和形狀估計(jì)-多場(chǎng)景約束的重要性
613.Monocular Relative Depth Perception With Web Stereo Data Supervision Web立體聲數(shù)據(jù)監(jiān)控的單眼相對(duì)深度感知
614.MorphNet: Fast & Simple Resource-Constrained Structure Learning of Deep Networks MorphNet：深度網(wǎng)絡(luò)的快速，簡(jiǎn)單的資源受限結(jié)構(gòu)學(xué)習(xí)
615.Motion-Appearance Co-Memory Networks for Video Question Answering 運(yùn)動(dòng)外觀協(xié)同存儲(chǔ)網(wǎng)絡(luò)，用于視頻問答
616.Motion-Guided Cascaded Refinement Network for Video Object Segmentation 運(yùn)動(dòng)引導(dǎo)級(jí)聯(lián)細(xì)化網(wǎng)絡(luò)的視頻對(duì)象分割
617.Motion Segmentation by Exploiting Complementary Geometric Models 利用互補(bǔ)幾何模型進(jìn)行運(yùn)動(dòng)分割
618.MovieGraphs: Towards Understanding Human-Centric Situations From Videos MovieGraphs：通過(guò)視頻了解以人為中心的情況
619.Multi-Agent Diverse Generative Adversarial Networks 多智能體多元化生成對(duì)抗網(wǎng)絡(luò)
620.Multi-Cell Detection and Classification Using a Generative Convolutional Model 使用生成卷積模型進(jìn)行多細(xì)胞檢測(cè)和分類
621.Multi-Content GAN for Few-Shot Font Style Transfer 多內(nèi)容GAN，可進(jìn)行少量字體轉(zhuǎn)換
622.Multi-Cue Correlation Filters for Robust Visual Tracking 多線索相關(guān)濾波器可實(shí)現(xiàn)強(qiáng)大的視覺跟蹤
623.Multi-Evidence Filtering and Fusion for Multi-Label Classification, Object Detection and Semantic Segmentation Based on Weakly Supervised Learning 基于弱監(jiān)督學(xué)習(xí)的多標(biāo)簽分類，目標(biāo)檢測(cè)和語(yǔ)義分割的多證據(jù)過(guò)濾與融合
624.Multi-Frame Quality Enhancement for Compressed Video 壓縮視頻的多幀質(zhì)量增強(qiáng)
625.Multi-Image Semantic Matching by Mining Consistent Features 挖掘一致特征的多圖像語(yǔ)義匹配
626.Multi-Label Zero-Shot Learning With Structured Knowledge Graphs 具有結(jié)構(gòu)化知識(shí)圖的多標(biāo)簽零射擊學(xué)習(xí)
627.Multi-Level Factorisation Net for Person Re-Identification 用于人員重新識(shí)別的多層次分解網(wǎng)絡(luò)
628.Multi-Level Fusion Based 3D Object Detection From Monocular Images 基于多級(jí)融合的單眼圖像3D目標(biāo)檢測(cè)
629.Multimodal Explanations: Justifying Decisions and Pointing to the Evidence 多式聯(lián)運(yùn)的解釋：做出合理的決定并指向證據(jù)
630.Multimodal Visual Concept Learning With Weakly Supervised Techniques 弱監(jiān)督技術(shù)的多模式視覺概念學(xué)習(xí)
631.Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation 通過(guò)角點(diǎn)定位和區(qū)域分割的多方位場(chǎng)景文本檢測(cè)
632.Multiple Granularity Group Interaction Prediction 多粒度群相互作用預(yù)測(cè)
633.Multi-Scale Location-Aware Kernel Representation for Object Detection 用于對(duì)象檢測(cè)的多尺度位置感知內(nèi)核表示
634.Multi-Scale Weighted Nuclear Norm Image Restoration 多尺度加權(quán)核規(guī)范圖像復(fù)原
635.Multi-Shot Pedestrian Re-Identification via Sequential Decision Making 通過(guò)順序決策進(jìn)行多步行人重新識(shí)別
636.Multispectral Image Intrinsic Decomposition via Subspace Constraint 通過(guò)子空間約束進(jìn)行多光譜圖像固有分解
637.Multistage Adversarial Losses for Pose-Based Human Image Synthesis 基于姿勢(shì)的人體圖像合成的多階段對(duì)抗性損失
638.Multi-Task Adversarial Network for Disentangled Feature Learning 多任務(wù)對(duì)抗網(wǎng)絡(luò)的融合特征學(xué)習(xí)
639.Multi-Task Learning by Maximizing Statistical Dependence 通過(guò)最大化統(tǒng)計(jì)依賴性進(jìn)行多任務(wù)學(xué)習(xí)
640.Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics 使用不確定性權(quán)衡場(chǎng)景幾何和語(yǔ)義損失的多任務(wù)學(xué)習(xí)
641.Multi-View Consistency as Supervisory Signal for Learning Shape and Pose Prediction 多視圖一致性作為學(xué)習(xí)形狀和姿勢(shì)預(yù)測(cè)的監(jiān)督信號(hào)
642.Multi-View Harmonized Bilinear Network for 3D Object Recognition 用于3D對(duì)象識(shí)別的多視圖協(xié)調(diào)雙線性網(wǎng)絡(luò)
643.MX-LSTM: Mixing Tracklets and Vislets to Jointly Forecast Trajectories and Head Poses MX-LSTM：將小軌跡和小片段混合以共同預(yù)測(cè)軌跡和頭部姿勢(shì)
644.NAG: Network for Adversary Generation NAG：對(duì)手生成網(wǎng)絡(luò)
645.Natural and Effective Obfuscation by Head Inpainting 通過(guò)頭部繪畫進(jìn)行自然而有效的混淆
646.NestedNet: Learning Nested Sparse Structures in Deep Neural Networks NestedNet：在深度神經(jīng)網(wǎng)絡(luò)中學(xué)習(xí)嵌套的稀疏結(jié)構(gòu)
647.Net2Vec: Quantifying and Explaining How Concepts Are Encoded by Filters in Deep Neural Networks Net2Vec：量化和解釋深度神經(jīng)網(wǎng)絡(luò)中的過(guò)濾器如何編碼概念
648.Neural 3D Mesh Renderer 神經(jīng)3D網(wǎng)格渲染器
649.Neural Baby Talk 神經(jīng)嬰兒談話
650.Neural Kinematic Networks for Unsupervised Motion Retargetting 神經(jīng)運(yùn)動(dòng)網(wǎng)絡(luò)的無(wú)監(jiān)督運(yùn)動(dòng)重定向

論文概要

651.Neural Motifs: Scene Graph Parsing With Global Context 神經(jīng)圖案：具有全局上下文的場(chǎng)景圖解析
652.NeuralNetwork-Viterbi: A Framework for Weakly Supervised Video Learning NeuralNetwork-Viterbi：弱監(jiān)督視頻學(xué)習(xí)的框架
653.Neural Sign Language Translation 神經(jīng)手語(yǔ)翻譯
654.Neural Style Transfer via Meta Networks 通過(guò)元網(wǎng)絡(luò)進(jìn)行神經(jīng)風(fēng)格傳遞
655.NISP: Pruning Networks Using Neuron Importance Score Propagation NISP：使用神經(jīng)元重要性分?jǐn)?shù)傳播修剪網(wǎng)絡(luò)
656.Non-Blind Deblurring: Handling Kernel Uncertainty With CNNs 非盲去模糊：使用CNN處理內(nèi)核不確定性
657.Nonlinear 3D Face Morphable Model 非線性3D人臉可變形模型
658.Non-Linear Temporal Subspace Representations for Activity Recognition 用于活動(dòng)識(shí)別的非線性時(shí)間子空間表示
659.Nonlocal Low-Rank Tensor Factor Analysis for Image Restoration 用于圖像復(fù)原的非局部低秩張量因子分析
660.Non-Local Neural Networks 非局部神經(jīng)網(wǎng)絡(luò)
661.Normalized Cut Loss for Weakly-Supervised CNN Segmentation 弱監(jiān)督CNN分割的歸一化割損
662.Now You Shake Me: Towards Automatic 4D Cinema 現(xiàn)在，您讓我搖了搖：邁向自動(dòng)4D電影院
663.OATM: Occlusion Aware Template Matching by Consensus Set Maximization OATM：通過(guò)共識(shí)集最大化來(lái)匹配遮擋感知模板
664.Object Referring in Videos With Language and Human Gaze 具有語(yǔ)言和人眼注視的視頻中的對(duì)象引用
665.Objects as Context for Detecting Their Semantic Parts 對(duì)象作為上下文來(lái)檢測(cè)其語(yǔ)義部分
666.Occluded Pedestrian Detection Through Guided Attention in CNNs 在CNN中通過(guò)引導(dǎo)注意力進(jìn)行行人檢測(cè)
667.Occlusion-Aware Rolling Shutter Rectification of 3D Scenes 遮擋感知型3D場(chǎng)景的卷簾式快門矯正
668.Occlusion Aware Unsupervised Learning of Optical Flow 遮擋感知光流的無(wú)監(jiān)督學(xué)習(xí)
669.OLE: Orthogonal Low-Rank Embedding - A Plug and Play Geometric Loss for Deep Learning OLE：正交低秩嵌入-深度學(xué)習(xí)的即插即用幾何損失
670.One-Shot Action Localization by Learning Sequence Matching Network 通過(guò)學(xué)習(xí)序列匹配網(wǎng)絡(luò)進(jìn)行一鍵式動(dòng)作定位
671.On the Convergence of PatchMatch and Its Variants 關(guān)于PatchMatch及其變體的收斂性
672.On the Duality Between Retinex and Image Dehazing Retinex與圖像去霧之間的對(duì)偶
673.On the Importance of Label Quality for Semantic Segmentation 標(biāo)簽質(zhì)量在語(yǔ)義分割中的重要性
674.On the Robustness of Semantic Segmentation Models to Adversarial Attacks 語(yǔ)義分割模型對(duì)對(duì)抗攻擊的魯棒性
675.Optical Flow Guided Feature: A Fast and Robust Motion Representation for Video Action Recognition 光流引導(dǎo)功能：用于視頻動(dòng)作識(shí)別的快速且魯棒的運(yùn)動(dòng)表示
676.Optimal Structured Light a La Carte 最佳結(jié)構(gòu)燈點(diǎn)菜
677.Optimizing Filter Size in Convolutional Neural Networks for Facial Action Unit Recognition 卷積神經(jīng)網(wǎng)絡(luò)中用于面部動(dòng)作單元識(shí)別的濾波器大小優(yōu)化
678.Optimizing Video Object Detection via a Scale-Time Lattice 通過(guò)比例時(shí)間格優(yōu)化視頻對(duì)象檢測(cè)
679.Ordinal Depth Supervision for 3D Human Pose Estimation 3D人體姿勢(shì)估計(jì)的序數(shù)深度監(jiān)督
680.PackNet: Adding Multiple Tasks to a Single Network by Iterative Pruning PackNet：通過(guò)迭代修剪將多個(gè)任務(wù)添加到單個(gè)網(wǎng)絡(luò)
681.PAD-Net: Multi-Tasks Guided Prediction-and-Distillation Network for Simultaneous Depth Estimation and Scene Parsing PAD-Net：多任務(wù)引導(dǎo)的預(yù)測(cè)和蒸餾網(wǎng)絡(luò)，用于同時(shí)深度估計(jì)和場(chǎng)景解析
682.PairedCycleGAN: Asymmetric Style Transfer for Applying and Removing Makeup PairedCycleGAN：不對(duì)稱樣式轉(zhuǎn)移，用于涂抹和去除化妝
683.Parallel Attention: A Unified Framework for Visual Object Discovery Through Dialogs and Queries 并行注意：通過(guò)對(duì)話框和查詢發(fā)現(xiàn)可視對(duì)象的統(tǒng)一框架
684.Partially Shared Multi-Task Convolutional Neural Network With Local Constraint for Face Attribute Learning 具有局部約束的部分共享多任務(wù)卷積神經(jīng)網(wǎng)絡(luò)用于人臉屬性學(xué)習(xí)
685.Partial Transfer Learning With Selective Adversarial Networks 選擇性對(duì)抗網(wǎng)絡(luò)的部分轉(zhuǎn)移學(xué)習(xí)
686.Path Aggregation Network for Instance Segmentation 用于實(shí)例分割的路徑聚合網(wǎng)絡(luò)
687.People, Penguins and Petri Dishes: Adapting Object Counting Models to New Visual Domains and Object Types Without Forgetting 人，企鵝和培養(yǎng)皿：在不忘記的情況下將對(duì)象計(jì)數(shù)模型適應(yīng)新的可視域和對(duì)象類型
688.Person Re-Identification With Cascaded Pairwise Convolutions 級(jí)聯(lián)成對(duì)卷積的人員重新識(shí)別
689.Person Transfer GAN to Bridge Domain Gap for Person Re-Identification 人員轉(zhuǎn)移GAN到橋接域差距以進(jìn)行人員重新識(shí)別
690.Perturbative Neural Networks 攝動(dòng)神經(jīng)網(wǎng)絡(luò)
691.PhaseNet for Video Frame Interpolation 用于視頻幀插值的PhaseNet
692.Photographic Text-to-Image Synthesis With a Hierarchically-Nested Adversarial Network 分層嵌套對(duì)抗網(wǎng)絡(luò)的攝影文本到圖像合成
693.Photometric Stereo in Participating Media Considering Shape-Dependent Forward Scatter 考慮形狀依賴性前向散射的參與介質(zhì)中的光度立體
694.PiCANet: Learning Pixel-Wise Contextual Attention for Saliency Detection PiCANet：學(xué)習(xí)像素性上下文注意以進(jìn)行顯著性檢測(cè)
695.PieAPP: Perceptual Image-Error Assessment Through Pairwise Preference PieAPP：通過(guò)成對(duì)偏好的感知圖像錯(cuò)誤評(píng)估
696.Pix3D: Dataset and Methods for Single-Image 3D Shape Modeling Pix3D：單圖像3D形狀建模的數(shù)據(jù)集和方法
697.Pixels, Voxels, and Views: A Study of Shape Representations for Single View 3D Object Shape Prediction 像素，體素和視圖：單視圖3D對(duì)象形狀預(yù)測(cè)的形狀表示研究
698.PIXOR: Real-Time 3D Object Detection From Point Clouds PIXOR：來(lái)自點(diǎn)云的實(shí)時(shí)3D對(duì)象檢測(cè)
699.Planar Shape Detection at Structural Scales 結(jié)構(gòu)尺度的平面形狀檢測(cè)
700.PlaneNet: Piece-Wise Planar Reconstruction From a Single RGB Image PlaneNet：從單個(gè)RGB圖像進(jìn)行明智的平面重建

論文概要

701.PointFusion: Deep Sensor Fusion for 3D Bounding Box Estimation PointFusion：用于3D邊界框估計(jì)的深度傳感器融合
702.PointGrid: A Deep Network for 3D Shape Understanding PointGrid：深入了解3D形狀的網(wǎng)絡(luò)
703.PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition PointNetVLAD：基于深度點(diǎn)云的大規(guī)模位置識(shí)別檢索
704.Pointwise Convolutional Neural Networks 點(diǎn)向卷積神經(jīng)網(wǎng)絡(luò)
705.Polarimetric Dense Monocular SLAM 偏振密集單眼SLAM
706.PoseFlow: A Deep Motion Representation for Understanding Human Behaviors in Videos PoseFlow：用于理解視頻中人類行為的深度運(yùn)動(dòng)表示
707.Pose-Guided Photorealistic Face Rotation 姿勢(shì)引導(dǎo)的真實(shí)感人臉旋轉(zhuǎn)
708.pOSE: Pseudo Object Space Error for Initialization-Free Bundle Adjustment pOSE：用于無(wú)初始化捆綁調(diào)整的偽對(duì)象空間錯(cuò)誤
709.Pose-Robust Face Recognition via Deep Residual Equivariant Mapping 基于深度殘差等變映射的姿勢(shì)魯棒人臉識(shí)別
710.PoseTrack: A Benchmark for Human Pose Estimation and Tracking PoseTrack：人體姿勢(shì)估計(jì)和跟蹤基準(zhǔn)
711.Pose Transferrable Person Re-Identification 姿勢(shì)可轉(zhuǎn)移人員的重新識(shí)別
712.PoTion: Pose MoTion Representation for Action Recognition PoTion：用于動(dòng)作識(shí)別的姿勢(shì)運(yùn)動(dòng)表示
713.PPFNet: Global Context Aware Local Features for Robust 3D Point Matching PPFNet：健壯的3D點(diǎn)匹配的全局上下文感知本地功能
714.Practical Block-Wise Neural Network Architecture Generation 實(shí)用的塊明智神經(jīng)網(wǎng)絡(luò)架構(gòu)生成
715.Preserving Semantic Relations for Zero-Shot Learning 保留語(yǔ)義關(guān)系以進(jìn)行零射擊學(xué)習(xí)
716.Probabilistic Joint Face-Skull Modelling for Facial Reconstruction 面部重建的概率聯(lián)合面顱骨模型
717.Probabilistic Plant Modeling via Multi-View Image-to-Image Translation 通過(guò)多視圖圖像到圖像轉(zhuǎn)換的概率植物建模
718.Progressive Attention Guided Recurrent Network for Salient Object Detection 漸進(jìn)式注意力引導(dǎo)循環(huán)網(wǎng)絡(luò)用于顯著物體檢測(cè)
719.Progressively Complementarity-Aware Fusion Network for RGB-D Salient Object Detection 用于RGB-D顯著目標(biāo)檢測(cè)的漸進(jìn)式互補(bǔ)感知融合網(wǎng)絡(luò)
720.Pseudo Mask Augmented Object Detection 偽蒙版增強(qiáng)對(duì)象檢測(cè)
721.Pulling Actions out of Context: Explicit Separation for Effective Combination 使動(dòng)作脫離上下文：有效組合的顯式分離
722.PU-Net: Point Cloud Upsampling Network PU-Net：點(diǎn)云上采樣網(wǎng)絡(luò)
723.PWC-Net: CNNs for Optical Flow Using Pyramid, Warping, and Cost Volume PWC-Net：使用金字塔，翹曲和成本量的光流CNN
724.Pyramid Stereo Matching Network 金字塔立體匹配網(wǎng)絡(luò)
725.Quantization and Training of Neural Networks for Efficient Integer-Arithmetic-Only Inference 神經(jīng)網(wǎng)絡(luò)的量化和訓(xùn)練，以便進(jìn)行有效的僅整數(shù)運(yùn)算
726.Quantization of Fully Convolutional Networks for Accurate Biomedical Image Segmentation 完全卷積網(wǎng)絡(luò)的量化，用于精確的生物醫(yī)學(xué)圖像分割
727.Radially-Distorted Conjugate Translations 徑向變形的共軛翻譯
728.RayNet: Learning Volumetric 3D Reconstruction With Ray Potentials RayNet：使用射線勢(shì)學(xué)習(xí)體積3D重建
729.Real-Time Monocular Depth Estimation Using Synthetic Data With Domain Adaptation via Image Style Transfer 通過(guò)圖像樣式轉(zhuǎn)移使用具有域自適應(yīng)的合成數(shù)據(jù)進(jìn)行實(shí)時(shí)單眼深度估計(jì)
730.Real-Time Rotation-Invariant Face Detection With Progressive Calibration Networks 漸進(jìn)式校準(zhǔn)網(wǎng)絡(luò)的實(shí)時(shí)旋轉(zhuǎn)不變?nèi)四槞z測(cè)
731.Real-Time Seamless Single Shot 6D Object Pose Prediction 實(shí)時(shí)無(wú)縫單發(fā)6D對(duì)象姿態(tài)預(yù)測(cè)
732.Real-World Anomaly Detection in Surveillance Videos 監(jiān)控視頻中的真實(shí)世界異常檢測(cè)
733.Real-World Repetition Estimation by Div, Grad and Curl 通過(guò)Div，Grad和Curl進(jìn)行真實(shí)世界的重復(fù)估計(jì)
734.Recognize Actions by Disentangling Components of Dynamics 通過(guò)解開動(dòng)力學(xué)的成分來(lái)識(shí)別動(dòng)作
735.Recognizing Human Actions as the Evolution of Pose Estimation Maps 將人類行為識(shí)別為姿勢(shì)估計(jì)圖的演變
736.Reconstructing Thin Structures of Manifold Surfaces by Integrating Spatial Curves 通過(guò)積分空間曲線重建歧管表面的薄結(jié)構(gòu)
737.Reconstruction Network for Video Captioning 視頻字幕重建網(wǎng)絡(luò)
738.Recovering Realistic Texture in Image Super-Resolution by Deep Spatial Feature Transform 通過(guò)深度空間特征變換在圖像超分辨率中恢復(fù)逼真的紋理
739.Recurrent Pixel Embedding for Instance Grouping 用于實(shí)例分組的遞歸像素嵌入
740.Recurrent Residual Module for Fast Inference in Videos 遞歸殘差模塊，可快速推斷視頻
741.Recurrent Saliency Transformation Network: Incorporating Multi-Stage Visual Cues for Small Organ Segmentation 循環(huán)顯著性轉(zhuǎn)換網(wǎng)絡(luò)：結(jié)合多階段視覺提示進(jìn)行小器官分割
742.Recurrent Scene Parsing With Perspective Understanding in the Loop 循環(huán)場(chǎng)景解析與透視理解
743.Recurrent Slice Networks for 3D Segmentation of Point Clouds 用于點(diǎn)云3D分割的遞歸切片網(wǎng)絡(luò)
744.Referring Image Segmentation via Recurrent Refinement Networks 通過(guò)遞歸細(xì)化網(wǎng)絡(luò)引用圖像分割
745.Referring Relationships 推薦關(guān)系
746.Reflection Removal for Large-Scale 3D Point Clouds 大型3D點(diǎn)云的反射消除
747.Regularizing Deep Networks by Modeling and Predicting Label Structure 通過(guò)建模和預(yù)測(cè)標(biāo)簽結(jié)構(gòu)來(lái)規(guī)范化深層網(wǎng)絡(luò)
748.Regularizing RNNs for Caption Generation by Reconstructing the Past With the Present 通過(guò)重建過(guò)去與現(xiàn)在來(lái)規(guī)范RNN以生成字幕
749.Reinforcement Cutting-Agent Learning for Video Object Segmentation 用于視頻對(duì)象分割的增強(qiáng)切割代理學(xué)習(xí)
750.Relation Networks for Object Detection 用于對(duì)象檢測(cè)的關(guān)系網(wǎng)絡(luò)

論文概要

751.Representing and Learning High Dimensional Data With the Optimal Transport Map From a Probabilistic Viewpoint 從概率角度用最佳傳輸圖表示和學(xué)習(xí)高維數(shù)據(jù)
752.Repulsion Loss: Detecting Pedestrians in a Crowd 排斥力損失：檢測(cè)人群中的行人
753.Residual Dense Network for Image Super-Resolution 殘留密集網(wǎng)絡(luò)可實(shí)現(xiàn)圖像超分辨率
754.Residual Parameter Transfer for Deep Domain Adaptation 深度域適應(yīng)的殘差參數(shù)傳遞
755.Resource Aware Person Re-Identification Across Multiple Resolutions 跨多種解決方案的資源感知人員重新識(shí)別
756.Rethinking Feature Distribution for Loss Functions in Image Classification 對(duì)圖像分類中損失函數(shù)的特征分布的重新思考
757.Rethinking the Faster R-CNN Architecture for Temporal Action Localization 重新思考用于時(shí)間動(dòng)作本地化的更快的R-CNN架構(gòu)
758.Revisiting Deep Intrinsic Image Decompositions 重新審視深度內(nèi)在圖像分解
759.Revisiting Dilated Convolution: A Simple Approach for Weakly- and Semi-Supervised Semantic Segmentation 重訪擴(kuò)張式卷積：一種用于弱監(jiān)督和半監(jiān)督語(yǔ)義分割的簡(jiǎn)單方法
760.Revisiting Knowledge Transfer for Training Object Class Detectors 復(fù)習(xí)訓(xùn)練對(duì)象類別檢測(cè)器的知識(shí)轉(zhuǎn)移
761.Revisiting Oxford and Paris: Large-Scale Image Retrieval Benchmarking 重溫牛津和巴黎：大型圖像檢索基準(zhǔn)
762.Revisiting Salient Object Detection: Simultaneous Detection, Ranking, and Subitizing of Multiple Salient Objects 回顧顯著對(duì)象檢測(cè)：多個(gè)顯著對(duì)象的同時(shí)檢測(cè)，排序和細(xì)分
763.Revisiting Video Saliency: A Large-Scale Benchmark and a New Model 回顧視頻顯著性：大規(guī)模基準(zhǔn)和新模型
764.Reward Learning From Narrated Demonstrations 敘述式學(xué)習(xí)中的獎(jiǎng)勵(lì)學(xué)習(xí)
765.Re-Weighted Adversarial Adaptation Network for Unsupervised Domain Adaptation 無(wú)權(quán)域適應(yīng)的加權(quán)加權(quán)對(duì)抗適應(yīng)網(wǎng)絡(luò)
766.R-FCN-3000 at 30fps: Decoupling Detection and Classification R-FCN-3000的30fps：解耦檢測(cè)和分類
767.Ring Loss: Convex Feature Normalization for Face Recognition 環(huán)損：用于面部識(shí)別的凸特征歸一化
768.ROAD: Reality Oriented Adaptation for Semantic Segmentation of Urban Scenes ROAD：針對(duì)城市場(chǎng)景的語(yǔ)義分割的面向現(xiàn)實(shí)的適應(yīng)
769.RoadTracer: Automatic Extraction of Road Networks From Aerial Images RoadTracer：從航拍圖像中自動(dòng)提取路網(wǎng)
770.Robust Classification With Convolutional Prototype Learning 卷積原型學(xué)習(xí)的魯棒分類
771.Robust Depth Estimation From Auto Bracketed Images 通過(guò)自動(dòng)包圍曝光圖像進(jìn)行穩(wěn)健的深度估計(jì)
772.Robust Facial Landmark Detection via a Fully-Convolutional Local-Global Context Network 通過(guò)全卷積局部全局上下文網(wǎng)絡(luò)進(jìn)行魯棒的面部地標(biāo)檢測(cè)
773.Robust Hough Transform Based 3D Reconstruction From Circular Light Fields 基于魯棒霍夫變換的圓形光場(chǎng)3D重構(gòu)
774.Robust Physical-World Attacks on Deep Learning Visual Classification 深度學(xué)習(xí)視覺分類的強(qiáng)大物理世界攻擊
775.Robust Video Content Alignment and Compensation for Rain Removal in a CNN Framework CNN框架中針對(duì)雨水去除的強(qiáng)大視頻內(nèi)容對(duì)齊和補(bǔ)償
776.Rolling Shutter and Radial Distortion Are Features for High Frame Rate Multi-Camera Tracking 滾動(dòng)快門和徑向失真是高幀率多攝像機(jī)跟蹤的功能
777.Rotation Averaging and Strong Duality 旋轉(zhuǎn)平均和強(qiáng)對(duì)偶
778.RotationNet: Joint Object Categorization and Pose Estimation Using Multiviews From Unsupervised Viewpoints RotationNet：使用來(lái)自無(wú)監(jiān)督觀點(diǎn)的多視圖進(jìn)行聯(lián)合對(duì)象分類和姿勢(shì)估計(jì)
779.Rotation-Sensitive Regression for Oriented Scene Text Detection 面向場(chǎng)景文本檢測(cè)的旋轉(zhuǎn)敏感回歸
780.Salience Guided Depth Calibration for Perceptually Optimized Compressive Light Field 3D Display 針對(duì)感知優(yōu)化的壓縮光場(chǎng)3D顯示的顯著性深度深度校準(zhǔn)
781.Salient Object Detection Driven by Fixation Prediction 固定預(yù)測(cè)驅(qū)動(dòng)的顯著物體檢測(cè)
782.SBNet: Sparse Blocks Network for Fast Inference SBNet：稀疏塊網(wǎng)絡(luò)以進(jìn)行快速推理
783.Scalable and Effective Deep CCA via Soft Decorrelation 通過(guò)軟解相關(guān)可擴(kuò)展且有效的深度CCA
784.Scalable Dense Non-Rigid Structure-From-Motion: A Grassmannian Perspective 運(yùn)動(dòng)可伸縮的密集非剛性結(jié)構(gòu)：格拉斯曼觀點(diǎn)
785.Scale-Recurrent Network for Deep Image Deblurring 用于深度圖像去模糊的縮放遞歸網(wǎng)絡(luò)
786.Scale-Transferrable Object Detection 可縮放的目標(biāo)檢測(cè)
787.ScanComplete: Large-Scale Scene Completion and Semantic Segmentation for 3D Scans ScanComplete：3D掃描的大規(guī)模場(chǎng)景完成和語(yǔ)義分割
788.SeedNet: Automatic Seed Generation With Deep Reinforcement Learning for Robust Interactive Segmentation SeedNet：具有深度強(qiáng)化學(xué)習(xí)功能的自動(dòng)種子生成，可實(shí)現(xiàn)可靠的交互式細(xì)分
789.Seeing Small Faces From Robust Anchor’s Perspective 從穩(wěn)固的錨點(diǎn)角度看小臉
790.Seeing Temporal Modulation of Lights From Standard Cameras 從標(biāo)準(zhǔn)相機(jī)看到光的時(shí)間調(diào)制
791.Seeing Voices and Hearing Faces: Cross-Modal Biometric Matching 看到聲音和聽覺的面孔：跨模態(tài)生物特征匹配
792.SeGAN: Segmenting and Generating the Invisible SeGAN：分割和生成不可見
793.Self-Calibrating Polarising Radiometric Calibration 自校準(zhǔn)偏振輻射校準(zhǔn)
794.Self-Supervised Adversarial Hashing Networks for Cross-Modal Retrieval 自我監(jiān)督的對(duì)抗式哈希網(wǎng)絡(luò)，用于跨模態(tài)檢索
795.Self-Supervised Feature Learning by Learning to Spot Artifacts 通過(guò)學(xué)習(xí)發(fā)現(xiàn)偽像進(jìn)行自我監(jiān)督的特征學(xué)習(xí)
796.Self-Supervised Learning of Geometrically Stable Features Through Probabilistic Introspection 通過(guò)概率自省對(duì)幾何穩(wěn)定特征進(jìn)行自我監(jiān)督學(xué)習(xí)
797.Self-Supervised Multi-Level Face Model Learning for Monocular Reconstruction at Over 250 Hz 用于250 Hz以上單眼重建的自監(jiān)督多級(jí)面部模型學(xué)習(xí)
798.Semantic Video Segmentation by Gated Recurrent Flow Propagation 門控循環(huán)流傳播的語(yǔ)義視頻分割
799.Semantic Visual Localization 語(yǔ)義視覺本地化
800.Semi-Parametric Image Synthesis 半?yún)?shù)圖像合成

論文概要

801.SemStyle: Learning to Generate Stylised Image Captions Using Unaligned Text SemStyle：學(xué)習(xí)使用未對(duì)齊的文本生成樣式化的圖像標(biāo)題
802.Separating Self-Expression and Visual Content in Hashtag Supervision 在標(biāo)簽監(jiān)督中分離自我表達(dá)和視覺內(nèi)容
803.Separating Style and Content for Generalized Style Transfer 分隔樣式和內(nèi)容以進(jìn)行廣義樣式轉(zhuǎn)移
804.SfSNet: Learning Shape, Reflectance and Illuminance of Faces `in the Wild’ SfSNet：學(xué)習(xí)“野外”面孔的形狀，反射率和照度
805.SGAN: An Alternative Training of Generative Adversarial Networks SGAN：生成對(duì)抗網(wǎng)絡(luò)的替代培訓(xùn)
806.SGPN: Similarity Group Proposal Network for 3D Point Cloud Instance Segmentation SGPN：3D點(diǎn)云實(shí)例細(xì)分的相似性組建議網(wǎng)絡(luò)
807.Shape From Shading Through Shape Evolution 從陰影到形狀進(jìn)化
808.Shift: A Zero FLOP, Zero Parameter Alternative to Spatial Convolutions 移位：零卷積，零參數(shù)替代空間卷積
809.Show Me a Story: Towards Coherent Neural Story Illustration 告訴我一個(gè)故事：邁向連貫的神經(jīng)故事插圖
810.ShuffleNet: An Extremely Efficient Convolutional Neural Network for Mobile Devices ShuffleNet：用于移動(dòng)設(shè)備的極其高效的卷積神經(jīng)網(wǎng)絡(luò)
811.Sim2Real Viewpoint Invariant Visual Servoing by Recurrent Control 通過(guò)遞歸控制實(shí)現(xiàn)Sim2Real Viewpoint不變視覺伺服
812.Single Image Dehazing via Conditional Generative Adversarial Network 通過(guò)條件生成對(duì)抗網(wǎng)絡(luò)進(jìn)行單圖像去霧
813.Single-Image Depth Estimation Based on Fourier Domain Analysis 基于傅立葉域分析的單圖像深度估計(jì)
814.Single Image Reflection Separation With Perceptual Losses 具有感知損失的單圖像反射分離
815.Single-Shot Object Detection With Enriched Semantics 具有豐富語(yǔ)義的單發(fā)目標(biāo)檢測(cè)
816.Single-Shot Refinement Neural Network for Object Detection 用于目標(biāo)檢測(cè)的單發(fā)細(xì)化神經(jīng)網(wǎng)絡(luò)
817.Single View Stereo Matching 單視圖立體聲匹配
818.SINT++: Robust Visual Tracking via Adversarial Positive Instance Generation SINT ++：通過(guò)對(duì)抗性積極實(shí)例生成進(jìn)行可靠的視覺跟蹤
819.Sketch-a-Classifier: Sketch-Based Photo Classifier Generation 草圖分類器：基于草圖的照片分類器生成
820.SketchMate: Deep Hashing for Million-Scale Human Sketch Retrieval SketchMate：數(shù)以百萬(wàn)計(jì)的人類草圖檢索的深度哈希
821.SketchyGAN: Towards Diverse and Realistic Sketch to Image Synthesis SketchyGAN：向圖像合成中逼真的素描
822.Sliced Wasserstein Distance for Learning Gaussian Mixture Models 切片Wasserstein距離用于學(xué)習(xí)高斯混合模型
823.Smooth Neighbors on Teacher Graphs for Semi-Supervised Learning 教師圖上的平滑鄰居用于半監(jiān)督學(xué)習(xí)
824.SobolevFusion: 3D Reconstruction of Scenes Undergoing Free Non-Rigid Motion SobolevFusion：進(jìn)行自由非剛性運(yùn)動(dòng)的場(chǎng)景的3D重建
825.Soccer on Your Tabletop 桌上足球
826.Social GAN: Socially Acceptable Trajectories With Generative Adversarial Networks 社交GAN：具有生成對(duì)抗網(wǎng)絡(luò)的社交可接受軌跡
827.Solving the Perspective-2-Point Problem for Flying-Camera Photo Composition 解決飛行相機(jī)照片構(gòu)圖的透視2點(diǎn)問題
828.SO-Net: Self-Organizing Network for Point Cloud Analysis SO-Net：用于點(diǎn)云分析的自組織網(wǎng)絡(luò)
829.SoS-RSC: A Sum-of-Squares Polynomial Approach to Robustifying Subspace Clustering Algorithms SoS-RSC：一種平方和多項(xiàng)式方法，用于魯棒子空間聚類算法
830.Sparse Photometric 3D Face Reconstruction Guided by Morphable Models 可變形模型指導(dǎo)的稀疏光度3D人臉重建
831.Sparse, Smart Contours to Represent and Edit Images 稀疏的智能輪廓來(lái)表示和編輯圖像
832.Spatially-Adaptive Filter Units for Deep Neural Networks 用于深度神經(jīng)網(wǎng)絡(luò)的空間自適應(yīng)濾波器單元
833.SPLATNet: Sparse Lattice Networks for Point Cloud Processing SPLATNet：用于點(diǎn)云處理的稀疏格子網(wǎng)絡(luò)
834.SplineCNN: Fast Geometric Deep Learning With Continuous B-Spline Kernels SplineCNN：具有連續(xù)B樣條曲線核的快速幾何深度學(xué)習(xí)
835.Spline Error Weighting for Robust Visual-Inertial Fusion 樣條誤差加權(quán)，實(shí)現(xiàn)穩(wěn)健的視覺慣性融合
836.Squeeze-and-Excitation Networks 擠壓和激勵(lì)網(wǎng)絡(luò)
837.SSNet: Scale Selection Network for Online 3D Action Prediction SSNet：在線3D動(dòng)作預(yù)測(cè)的量表選擇網(wǎng)絡(luò)
838.Stacked Conditional Generative Adversarial Networks for Jointly Learning Shadow Detection and Shadow Removal 聯(lián)合學(xué)習(xí)陰影檢測(cè)和陰影去除的堆疊條件生成對(duì)抗網(wǎng)絡(luò)
839.Stacked Latent Attention for Multimodal Reasoning 多模態(tài)推理的堆疊潛在注意力
840.StarGAN: Unified Generative Adversarial Networks for Multi-Domain Image-to-Image Translation StarGAN：用于多域圖像到圖像翻譯的統(tǒng)一生成對(duì)抗網(wǎng)絡(luò)
841.Statistical Tomography of Microscopic Life 微觀生命的統(tǒng)計(jì)斷層掃描
842.Stereoscopic Neural Style Transfer 立體神經(jīng)風(fēng)格轉(zhuǎn)換
843.ST-GAN: Spatial Transformer Generative Adversarial Networks for Image Compositing ST-GAN：用于圖像合成的空間變壓器生成對(duì)抗網(wǎng)絡(luò)
844.Stochastic Downsampling for Cost-Adjustable Inference and Improved Regularization in Convolutional Networks 卷積網(wǎng)絡(luò)中用于成本可調(diào)推斷和改進(jìn)正則化的隨機(jī)下采樣
845.Stochastic Variational Inference With Gradient Linearization 梯度線性化的隨機(jī)變分推斷
846.Structured Attention Guided Convolutional Neural Fields for Monocular Depth Estimation 結(jié)構(gòu)化注意力導(dǎo)向的卷積神經(jīng)場(chǎng)用于單眼深度估計(jì)
847.Structured Set Matching Networks for One-Shot Part Labeling 一站式零件貼標(biāo)的結(jié)構(gòu)化集合匹配網(wǎng)絡(luò)
848.Structured Uncertainty Prediction Networks 結(jié)構(gòu)化不確定性預(yù)測(cè)網(wǎng)絡(luò)
849.Structure From Recurrent Motion: From Rigidity to Recurrency 循環(huán)運(yùn)動(dòng)的結(jié)構(gòu)：從剛度到遞歸
850.Structure Inference Net: Object Detection Using Scene-Level Context and Instance-Level Relationships 結(jié)構(gòu)推斷網(wǎng)：使用場(chǎng)景級(jí)上下文和實(shí)例級(jí)關(guān)系的對(duì)象檢測(cè)

論文概要

851.Structure Preserving Video Prediction 保留結(jié)構(gòu)的視頻預(yù)測(cè)
852.Style Aggregated Network for Facial Landmark Detection 用于面部地標(biāo)檢測(cè)的樣式聚合網(wǎng)絡(luò)
853.Super-FAN: Integrated Facial Landmark Localization and Super-Resolution of Real-World Low Resolution Faces in Arbitrary Poses With GANs Super-FAN：具有GAN的任意姿勢(shì)中的集成面部地標(biāo)定位和真實(shí)世界低分辨率面孔的超分辨率
854.Super-Resolving Very Low-Resolution Face Images With Supplementary Attributes 具有補(bǔ)充屬性的超分辨率超低分辨率人臉圖像
855.Super SloMo: High Quality Estimation of Multiple Intermediate Frames for Video Interpolation Super SloMo：用于視頻插值的多個(gè)中間幀的高質(zhì)量估計(jì)
856.Supervision-by-Registration: An Unsupervised Approach to Improve the Precision of Facial Landmark Detectors 注冊(cè)監(jiān)督：一種無(wú)監(jiān)督的方法來(lái)提高面部地標(biāo)檢測(cè)器的精度
857.Surface Networks 地面網(wǎng)絡(luò)
858.SurfConv: Bridging 3D and 2D Convolution for RGBD Images SurfConv：橋接RGBD圖像的3D和2D卷積
859.Synthesizing Images of Humans in Unseen Poses 在看不見的姿勢(shì)中合成人的形象
860.SYQ: Learning Symmetric Quantization for Efficient Deep Neural Networks SYQ：為高效的深度神經(jīng)網(wǎng)絡(luò)學(xué)習(xí)對(duì)稱量化
861.Tagging Like Humans: Diverse and Distinct Image Annotation 像人類一樣標(biāo)記：多樣化且獨(dú)特的圖像注釋
862.Tags2Parts: Discovering Semantic Regions From Shape Tags 標(biāo)簽2部分：從形狀標(biāo)簽中發(fā)現(xiàn)語(yǔ)義區(qū)域
863.Tangent Convolutions for Dense Prediction in 3D 切線卷積用于3D密集預(yù)測(cè)
864.Taskonomy: Disentangling Task Transfer Learning Taskonomy：解開任務(wù)轉(zhuǎn)移學(xué)習(xí)
865.Teaching Categories to Human Learners With Visual Explanations 用視覺解釋向人類學(xué)習(xí)者教授類別
866.Tell Me Where to Look: Guided Attention Inference Network 告訴我在哪里看：引導(dǎo)注意推理網(wǎng)絡(luò)
867.Temporal Deformable Residual Networks for Action Segmentation in Videos 視頻中的時(shí)間分段的時(shí)間可變形殘差網(wǎng)絡(luò)
868.Temporal Hallucinating for Action Recognition With Few Still Images 時(shí)間幻覺的動(dòng)作識(shí)別很少有靜止圖像
869.Tensorize, Factorize and Regularize: Robust Visual Relationship Learning 張量化，分解和正則化：穩(wěn)健的視覺關(guān)系學(xué)習(xí)
870.Textbook Question Answering Under Instructor Guidance With Memory Networks 記憶網(wǎng)絡(luò)下教師指導(dǎo)下的教科書問答
871.TextureGAN: Controlling Deep Image Synthesis With Texture Patches TextureGAN：使用紋理補(bǔ)丁控制深度圖像合成
872.Texture Mapping for 3D Reconstruction With RGB-D Sensor 使用RGB-D傳感器進(jìn)行3D重建的紋理映射
873.The Best of Both Worlds: Combining CNNs and Geometric Constraints for Hierarchical Motion Segmentation 兩全其美：結(jié)合CNN和幾何約束進(jìn)行分層運(yùn)動(dòng)分割
874.The INaturalist Species Classification and Detection Dataset 非自然物種分類和檢測(cè)數(shù)據(jù)集
875.The LovaSz-Softmax Loss: A Tractable Surrogate for the Optimization of the Intersection-Over-Union Measure in Neural Networks LovaSz-Softmax損失：神經(jīng)網(wǎng)絡(luò)交叉口聯(lián)合測(cè)量的優(yōu)化的可替代替代方法
876.The Perception-Distortion Tradeoff 感知失真權(quán)衡
877.The Power of Ensembles for Active Learning in Image Classification 集成在圖像分類中主動(dòng)學(xué)習(xí)的力量
878.The Unreasonable Effectiveness of Deep Features as a Perceptual Metric 深度特征作為感知指標(biāo)的不合理有效性
879.Thoracic Disease Identification and Localization With Limited Supervision 有限監(jiān)督下的胸腔疾病鑒定和定位
880.Through-Wall Human Pose Estimation Using Radio Signals 使用無(wú)線電信號(hào)的全程人體姿態(tài)估計(jì)
881.TieNet: Text-Image Embedding Network for Common Thorax Disease Classification and Reporting in Chest X-Rays TieNet：普通胸部疾病分類和胸部X光報(bào)告的文本圖像嵌入網(wǎng)絡(luò)
882.Time-Resolved Light Transport Decomposition for Thermal Photometric Stereo 熱光度立體的時(shí)間分辨光傳輸分解
883.Tips and Tricks for Visual Question Answering: Learnings From the 2017 Challenge 視覺問題解答的提示和技巧：2017年挑戰(zhàn)的經(jīng)驗(yàn)教訓(xùn)
884.TOM-Net: Learning Transparent Object Matting From a Single Image TOM-Net：從單個(gè)圖像學(xué)習(xí)透明對(duì)象遮罩
885.Total Capture: A 3D Deformation Model for Tracking Faces, Hands, and Bodies 總捕獲：用于跟蹤面部，手部和身體的3D變形模型
886.Toward Driving Scene Understanding: A Dataset for Learning Driver Behavior and Causal Reasoning 對(duì)駕駛場(chǎng)景的理解：學(xué)習(xí)駕駛員行為和因果推理的數(shù)據(jù)集
887.Towards a Mathematical Understanding of the Difficulty in Learning With Feedforward Neural Networks 對(duì)前饋神經(jīng)網(wǎng)絡(luò)學(xué)習(xí)困難的數(shù)學(xué)理解
888.Towards Dense Object Tracking in a 2D Honeybee Hive 在2D蜜蜂蜂巢中實(shí)現(xiàn)密集對(duì)象跟蹤
889.Towards Effective Low-Bitwidth Convolutional Neural Networks 邁向有效的低位寬卷積神經(jīng)網(wǎng)絡(luò)
890.Towards Faster Training of Global Covariance Pooling Networks by Iterative Matrix Square Root Normalization 通過(guò)迭代矩陣平方根歸一化來(lái)更快地訓(xùn)練全局協(xié)方差合并網(wǎng)絡(luò)
891.Towards High Performance Video Object Detection 邁向高性能視頻目標(biāo)檢測(cè)
892.Towards Human-Machine Cooperation: Self-Supervised Sample Mining for Object Detection 走向人機(jī)合作：用于對(duì)象檢測(cè)的自監(jiān)督樣本挖掘
893.Towards Open-Set Identity Preserving Face Synthesis 邁向開放式身份保留人臉綜合
894.Towards Pose Invariant Face Recognition in the Wild 走向野外姿勢(shì)不變的人臉識(shí)別
895.Towards Universal Representation for Unseen Action Recognition 走向通用表示以實(shí)現(xiàn)看不見的動(dòng)作識(shí)別
896.Tracking Multiple Objects Outside the Line of Sight Using Speckle Imaging 使用散斑成像跟蹤視線外的多個(gè)物體
897.Transductive Unbiased Embedding for Zero-Shot Learning 零射學(xué)習(xí)的傳導(dǎo)無(wú)偏嵌入
898.Transferable Joint Attribute-Identity Deep Learning for Unsupervised Person Re-Identification 可轉(zhuǎn)移的聯(lián)合屬性-身份深度學(xué)習(xí)，用于無(wú)監(jiān)督人員重新識(shí)別
899.Translating and Segmenting Multimodal Medical Volumes With Cycle- and Shape-Consistency Generative Adversarial Network 使用周期和形狀一致性生成對(duì)抗網(wǎng)絡(luò)對(duì)多峰醫(yī)療量進(jìn)行翻譯和分段
900.Transparency by Design: Closing the Gap Between Performance and Interpretability in Visual Reasoning 設(shè)計(jì)上的透明度：彌合視覺推理中性能和可解釋性之間的差距

論文概要

901.Trapping Light for Time of Flight 飛行時(shí)間的陷阱燈
902.Triplet-Center Loss for Multi-View 3D Object Retrieval 多視圖3D對(duì)象檢索的三重態(tài)中心損失
903.Trust Your Model: Light Field Depth Estimation With Inline Occlusion Handling 信任您的模型：內(nèi)聯(lián)遮擋處理的光場(chǎng)深度估計(jì)
904.Two Can Play This Game: Visual Dialog With Discriminative Question Generation and Answering 兩個(gè)人可以玩這個(gè)游戲：具有判別性問題生成和回答的可視對(duì)話框
905.Two-Step Quantization for Low-Bit Neural Networks 低位神經(jīng)網(wǎng)絡(luò)的兩步量化
906.Two-Stream Convolutional Networks for Dynamic Texture Synthesis 用于動(dòng)態(tài)紋理合成的兩流卷積網(wǎng)絡(luò)
907.Uncalibrated Photometric Stereo Under Natural Illumination 自然照明下未經(jīng)校準(zhǔn)的測(cè)光立體
908.Unifying Identification and Context Learning for Person Recognition 統(tǒng)一身份識(shí)別和上下文學(xué)習(xí)以實(shí)現(xiàn)人的識(shí)別
909.Universal Denoising Networks : A Novel CNN Architecture for Image Denoising 通用降噪網(wǎng)絡(luò)：一種用于圖像降噪的新型CNN架構(gòu)
910.Unsupervised Correlation Analysis 無(wú)監(jiān)督相關(guān)分析
911.Unsupervised Cross-Dataset Person Re-Identification by Transfer Learning of Spatial-Temporal Patterns 通過(guò)時(shí)空模式的轉(zhuǎn)移學(xué)習(xí)對(duì)無(wú)監(jiān)督的跨數(shù)據(jù)集人員進(jìn)行重新識(shí)別
912.Unsupervised Deep Generative Adversarial Hashing Network 無(wú)監(jiān)督的深度生成對(duì)抗式哈希網(wǎng)絡(luò)
913.Unsupervised Discovery of Object Landmarks as Structural Representations 無(wú)監(jiān)督地發(fā)現(xiàn)對(duì)象地標(biāo)作為結(jié)構(gòu)表示形式
914.Unsupervised Domain Adaptation With Similarity Learning 具有相似性學(xué)習(xí)的無(wú)監(jiān)督域自適應(yīng)
915.Unsupervised Feature Learning via Non-Parametric Instance Discrimination 通過(guò)非參數(shù)實(shí)例區(qū)分進(jìn)行無(wú)監(jiān)督特征學(xué)習(xí)
916.Unsupervised Learning and Segmentation of Complex Activities From Video 視頻的無(wú)監(jiān)督學(xué)習(xí)和復(fù)雜活動(dòng)細(xì)分
917.Unsupervised Learning of Depth and Ego-Motion From Monocular Video Using 3D Geometric Constraints 使用3D幾何約束從單眼視頻進(jìn)行無(wú)監(jiān)督的深度和自我運(yùn)動(dòng)學(xué)習(xí)
918.Unsupervised Learning of Monocular Depth Estimation and Visual Odometry With Deep Feature Reconstruction 具有深度特征重構(gòu)的單眼深度估計(jì)和視覺測(cè)程的無(wú)監(jiān)督學(xué)習(xí)
919.Unsupervised Person Image Synthesis in Arbitrary Poses 任意姿勢(shì)下的無(wú)人監(jiān)督圖像合成
920.Unsupervised Sparse Dirichlet-Net for Hyperspectral Image Super-Resolution 無(wú)監(jiān)督稀疏Dirichlet-Net用于高光譜圖像超分辨率
921.Unsupervised Textual Grounding: Linking Words to Image Concepts 無(wú)監(jiān)督的文本基礎(chǔ)：將單詞鏈接到圖像概念
922.Unsupervised Training for 3D Morphable Model Regression 3D變形模型回歸的無(wú)監(jiān)督訓(xùn)練
923.UV-GAN: Adversarial Facial UV Map Completion for Pose-Invariant Face Recognition UV-GAN：對(duì)抗性面部UV貼圖完成，用于姿勢(shì)不變的面部識(shí)別
924.V2V-PoseNet: Voxel-to-Voxel Prediction Network for Accurate 3D Hand and Human Pose Estimation From a Single Depth Map V2V-PoseNet：用于從單個(gè)深度圖進(jìn)行精確3D手和人姿估計(jì)的體素到體素預(yù)測(cè)網(wǎng)絡(luò)
925.Variational Autoencoders for Deforming 3D Mesh Models 變形3D網(wǎng)格模型的變體自動(dòng)編碼器
926.Very Large-Scale Global SfM by Distributed Motion Averaging 分布式運(yùn)動(dòng)平均的超大規(guī)模全球SfM
927.Video Based Reconstruction of 3D People Models 基于視頻的3D人模型重構(gòu)
928.Video Captioning via Hierarchical Reinforcement Learning 通過(guò)分層強(qiáng)化學(xué)習(xí)進(jìn)行視頻字幕
929.Video Person Re-Identification With Competitive Snippet-Similarity Aggregation and Co-Attentive Snippet Embedding 具有競(jìng)爭(zhēng)性摘要相似性聚合和共同關(guān)注性摘要嵌入的視頻人重新識(shí)別
930.Video Rain Streak Removal by Multiscale Convolutional Sparse Coding 多尺度卷積稀疏編碼去除視頻雨紋
931.Video Representation Learning Using Discriminative Pooling 使用區(qū)分池的視頻表示學(xué)習(xí)
932.View Extrapolation of Human Body From a Single Image 從單個(gè)圖像查看人體外推
933.Viewpoint-Aware Attentive Multi-View Inference for Vehicle Re-Identification 用于車輛重新識(shí)別的具有視點(diǎn)意識(shí)的多視圖推理
934.Viewpoint-Aware Video Summarization 視點(diǎn)感知視頻摘要
935.VirtualHome: Simulating Household Activities via Programs VirtualHome：通過(guò)程序模擬家庭活動(dòng)
936.Vision-and-Language Navigation: Interpreting Visually-Grounded Navigation Instructions in Real Environments 視覺和語(yǔ)言導(dǎo)航：解釋真實(shí)環(huán)境中的視覺地面導(dǎo)航說(shuō)明
937.Visual Feature Attribution Using Wasserstein GANs 使用Wasserstein GAN的視覺特征歸因
938.Visual Grounding via Accumulated Attention 通過(guò)累積注意力進(jìn)行視覺接地
939.Visual Question Answering With Memory-Augmented Networks 內(nèi)存增強(qiáng)網(wǎng)絡(luò)的視覺問題解答
940.Visual Question Generation as Dual Task of Visual Question Answering 視覺問題生成是視覺問題回答的雙重任務(wù)
941.Visual Question Reasoning on General Dependency Tree 一般依賴樹上的視覺問題推理
942.Visual to Sound: Generating Natural Sound for Videos in the Wild 視覺到聲音：為野外視頻生成自然聲音
943.VITAL: VIsual Tracking via Adversarial Learning VITAL：通過(guò)對(duì)抗性學(xué)習(xí)進(jìn)行視覺跟蹤
944.VITON: An Image-Based Virtual Try-On Network VITON：基于映像的虛擬試穿網(wǎng)絡(luò)
945.VizWiz Grand Challenge: Answering Visual Questions From Blind People VizWiz大挑戰(zhàn)：回答盲人的視覺問題
946.VoxelNet: End-to-End Learning for Point Cloud Based 3D Object Detection VoxelNet：基于點(diǎn)云的3D對(duì)象檢測(cè)的端到端學(xué)習(xí)
947.W2F: A Weakly-Supervised to Fully-Supervised Framework for Object Detection W2F：弱監(jiān)督到完全監(jiān)督的對(duì)象檢測(cè)框架
948.Wasserstein Introspective Neural Networks Wasserstein自省神經(jīng)網(wǎng)絡(luò)
949.Weakly and Semi Supervised Human Body Part Parsing via Pose-Guided Knowledge Transfer 通過(guò)姿勢(shì)指導(dǎo)的知識(shí)轉(zhuǎn)移進(jìn)行弱而半監(jiān)督的人體部位解析
950.Weakly Supervised Action Localization by Sparse Temporal Pooling Network 稀疏時(shí)間池網(wǎng)絡(luò)對(duì)行為的弱監(jiān)督

論文概要

951.Weakly-Supervised Action Segmentation With Iterative Soft Boundary Assignment 具有迭代軟邊界分配的弱監(jiān)督動(dòng)作細(xì)分
952.Weakly Supervised Coupled Networks for Visual Sentiment Analysis 弱監(jiān)督耦合網(wǎng)絡(luò)，用于視覺情感分析
953.Weakly-Supervised Deep Convolutional Neural Network Learning for Facial Action Unit Intensity Estimation 弱監(jiān)督深度卷積神經(jīng)網(wǎng)絡(luò)學(xué)習(xí)，用于面部動(dòng)作單元強(qiáng)度估計(jì)
954.Weakly Supervised Facial Action Unit Recognition Through Adversarial Training 通過(guò)對(duì)抗訓(xùn)練對(duì)面部動(dòng)作單元識(shí)別進(jìn)行監(jiān)督不足
955.Weakly Supervised Instance Segmentation Using Class Peak Response 使用類峰值響應(yīng)的弱監(jiān)督實(shí)例分割
956.Weakly Supervised Learning of Single-Cell Feature Embeddings 單細(xì)胞特征嵌入的弱監(jiān)督學(xué)習(xí)
957.Weakly Supervised Phrase Localization With Multi-Scale Anchored Transformer Network 多尺度錨定變壓器網(wǎng)絡(luò)的弱監(jiān)督短語(yǔ)定位
958.Weakly-Supervised Semantic Segmentation by Iteratively Mining Common Object Features 通過(guò)迭代挖掘常見對(duì)象特征的弱監(jiān)督語(yǔ)義分割
959.Weakly-Supervised Semantic Segmentation Network With Deep Seeded Region Growing 具有深度種子區(qū)域增長(zhǎng)的弱監(jiān)督語(yǔ)義分割網(wǎng)絡(luò)
960.Webly Supervised Learning Meets Zero-Shot Learning: A Hybrid Approach for Fine-Grained Classification 網(wǎng)上監(jiān)督學(xué)習(xí)與零射擊學(xué)習(xí)：精細(xì)分類的混合方法
961.What Do Deep Networks Like to See? 深度網(wǎng)絡(luò)喜歡看什么？
962.What Have We Learned From Deep Representations for Action Recognition? 我們從用于動(dòng)作識(shí)別的深度表示中學(xué)到了什么？
963.What Makes a Video a Video: Analyzing Temporal Information in Video Understanding Models and Datasets 是什么使視頻成為視頻：分析視頻中的時(shí)間信息了解模型和數(shù)據(jù)集
964.When Will You Do What? - Anticipating Temporal Occurrences of Activities 您什么時(shí)候會(huì)做什么？ -預(yù)期活動(dòng)的臨時(shí)發(fā)生
965.Where and Why Are They Looking? Jointly Inferring Human Attention and Intentions in Complex Tasks 他們?cè)谀睦锟?#xff0c;為什么看？共同推斷人類的注意力和復(fù)雜任務(wù)中的意圖
966.Who Let the Dogs Out? Modeling Dog Behavior From Visual Data 誰(shuí)讓狗出去了？根據(jù)視覺數(shù)據(jù)模擬狗的行為
967.Who’s Better? Who’s Best? Pairwise Deep Ranking for Skill Determination 誰(shuí)更好？誰(shuí)最好成對(duì)確定技能的深度排名
968.Wide Compression: Tensor Ring Nets 寬壓縮：張量環(huán)網(wǎng)
969.WILDTRACK: A Multi-Camera HD Dataset for Dense Unscripted Pedestrian Detection WILDTRACK：用于密集無(wú)腳本行人檢測(cè)的多攝像機(jī)高清數(shù)據(jù)集
970.Wing Loss for Robust Facial Landmark Localisation With Convolutional Neural Networks 用卷積神經(jīng)網(wǎng)絡(luò)進(jìn)行穩(wěn)健的人臉地標(biāo)定位的機(jī)翼?yè)p失
971.Wrapped Gaussian Process Regression on Riemannian Manifolds 黎曼流形上的包裹高斯過(guò)程回歸
972.xUnit: Learning a Spatial Activation Function for Efficient Image Restoration xUnit：學(xué)習(xí)空間激活功能以進(jìn)行有效的圖像恢復(fù)
973.Zero-Shot Kernel Learning 零射內(nèi)核學(xué)習(xí)
974.Zero-Shot Recognition via Semantic Embeddings and Knowledge Graphs 通過(guò)語(yǔ)義嵌入和知識(shí)圖進(jìn)行零散識(shí)別
975.Zero-Shot Sketch-Image Hashing 零射素描圖像散列
976.“Zero-Shot” Super-Resolution Using Deep Internal Learning 使用深度內(nèi)部學(xué)習(xí)的“零射擊”超分辨率
977.Zero-Shot Visual Recognition Using Semantics-Preserving Adversarial Embedding Networks 保留語(yǔ)義對(duì)抗性嵌入網(wǎng)絡(luò)的零射視覺識(shí)別
978.Zigzag Learning for Weakly Supervised Object Detection 鋸齒形學(xué)習(xí)用于弱監(jiān)督對(duì)象檢測(cè)
979.Zoom and Learn: Generalizing Deep Stereo Matching to Novel Domains 縮放和學(xué)習(xí)：將深度立體聲匹配推廣到新穎領(lǐng)域

總結(jié)

以上是生活随笔為你收集整理的（十：2020.08.28）CVPR 2018 追踪之论文纲要（译）的全部?jī)?nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯(cuò)，歡迎將生活随笔推薦給好友。