ECCV 2020 accepted papers on Video
152 of the 1,358 papers accepted at ECCV 2020 (11.2%) are on Video, compared with 13.4% at ECCV 2018. They fall into 4 topics.
See ECCV 2020 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Video segmentation and tracking · 72 papers
- Two-branch Recurrent Network for Isolating Deepfakes in Videos (494 citations)
- Learning Joint Spatial-Temporal Transformations for Video Inpainting (321 citations)
- Video Object Segmentation with Episodic Graph Memory Networks (292 citations)
- Clustering Driven Deep Autoencoder for Video Anomaly Detection (283 citations)
- Video Super-Resolution with Recurrent Structure-Detail Network (270 citations)
- Multi-Temporal Recurrent Neural Networks For Progressive Non-Uniform Single Image Deblurring With Incremental Temporal Training (269 citations)
- Collaborative Video Object Segmentation by Foreground-Background Integration (265 citations)
- Self-Supervised Learning of Audio-Visual Objects from Video (233 citations)
- BMBC: Bilateral Motion Estimation with Bilateral Cost Volume for Video Interpolation (227 citations)
- Fully Convolutional Networks for Continuous Sign Language Recognition (215 citations)
- Reducing the Sim-to-Real Gap for Event Cameras (212 citations)
- Kernelized Memory Network for Video Object Segmentation (210 citations)
and 60 more in the interactive map.
Action recognition · 45 papers
- Gait Lateral Network: Learning Discriminative and Compact Representations for Gait Recognition (240 citations)
- Self-supervised Video Representation Learning by Pace Prediction (227 citations)
- Memory-augmented Dense Predictive Coding for Video Representation Learning (217 citations)
- Few-shot Action Recognition with Permutation-invariant Attention (208 citations)
- VPN: Learning Video-Pose Embedding for Activities of Daily Living (201 citations)
- Bottom-Up Temporal Action Localization with Mutual Regularization (187 citations)
- AR-Net: Adaptive Frame Resolution for Efficient Action Recognition (167 citations)
- Boundary Content Graph Neural Network for Temporal Action Proposal Generation (161 citations)
- Two-Stream Consensus Network for Weakly-Supervised Temporal Action Localization (158 citations)
- SF-Net: Single-Frame Supervision for Temporal Action Localization (156 citations)
- MotionSqueeze: Neural Motion Feature Learning for Video Understanding (153 citations)
- Boundary-Aware Cascade Networks for Temporal Action Segmentation (152 citations)
and 33 more in the interactive map.
Video-language understanding · 19 papers
- Multi-modal Transformer for Video Retrieval (631 citations)
- TVR: A Large-Scale Dataset for Video-Subtitle Moment Retrieval (294 citations)
- MovieNet: A Holistic Dataset for Movie Understanding (196 citations)
- Learning Modality Interaction for Temporal Sentence Localization and Event Captioning in Videos (100 citations)
- VLANet: Video-Language Alignment Network for Weakly-Supervised Video Moment Retrieval (91 citations)
- Global-and-Local Relative Position Embedding for Unsupervised Video Summarization (65 citations)
- SODA: Story Oriented Dense Video Captioning Evaluation Framework (58 citations)
- Traffic Accident Benchmark for Causality Recognition (55 citations)
- Visual Relation Grounding in Videos (46 citations)
- Hierarchical Visual-Textual Graph for Temporal Activity Localization via Language (43 citations)
- Knowledge-Based Video Question Answering with Unsupervised Scene Descriptions (40 citations)
- Multimodal Memorability: Modeling Effects of Semantics and Decay on Video Memorability (40 citations)
and 7 more in the interactive map.
Video generation · 16 papers
- Neural Voice Puppetry: Audio-driven Facial Reenactment (409 citations)
- MEAD: A Large-scale Audio-visual Dataset for Emotional Talking-face Generation (375 citations)
- DLow: Diversifying Latent Flows for Diverse Human Motion Prediction (276 citations)
- Unified Image and Video Saliency Modeling (205 citations)
- Talking-head Generation with Rhythmic Head Motion (192 citations)
- Foley Music: Learning to Generate Music from Videos (140 citations)
- Speech-driven Facial Animation using Cascaded GANs for Learning of Motion and Texture (105 citations)
- World-Consistent Video-to-Video Synthesis (90 citations)
- Character-Preserving Coherent Story Visualization (41 citations)
- Hierarchical Style-based Networks for Motion Synthesis (30 citations)
- Sound2Sight: Generating Visual Dynamics from Sound and Context (28 citations)
- Identity-Aware Multi-Sentence Video Description (19 citations)
and 4 more in the interactive map.
