CVPR 2021 accepted papers on Video
170 of the 1,661 papers accepted at CVPR 2021 (10.2%) are on Video, compared with 8.1% at CVPR 2020. They fall into 4 topics.
See CVPR 2021 in the interactive map
Submitting to CVPR? Checklist →
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 · 93 papers
- End-to-End Video Instance Segmentation With Transformers (654 citations)
- BasicVSR: The Search for Essential Components in Video Super-Resolution and Beyond (615 citations)
- Lips Don't Lie: A Generalisable and Robust Approach To Face Forgery Detection (501 citations)
- Multimodal Motion Prediction With Stacked Transformers (475 citations)
- Spatiotemporal Contrastive Video Representation Learning (416 citations)
- MIST: Multiple Instance Self-Training Framework for Video Anomaly Detection (349 citations)
- FVC: A New Framework Towards Deep Video Compression in Feature Space (273 citations)
- VideoMoCo: Contrastive Video Representation Learning With Temporally Adversarial Examples (229 citations)
- Time Lens: Event-Based Video Frame Interpolation (227 citations)
- Learning Normal Dynamics in Videos With Meta Prototype Network (222 citations)
- A Large-Scale Study on Unsupervised Spatiotemporal Representation Learning (218 citations)
- MotionRNN: A Flexible Model for Video Prediction With Spacetime-Varying Motions (205 citations)
and 81 more in the interactive map.
Action recognition · 50 papers
- TDN: Temporal Difference Networks for Efficient Action Recognition (489 citations)
- Learning Salient Boundary Feature for Anchor-free Temporal Action Localization (288 citations)
- UAV-Human: A Large Benchmark for Human Behavior Understanding With Unmanned Aerial Vehicles (277 citations)
- MoViNets: Mobile Video Networks for Efficient Video Recognition (252 citations)
- ACTION-Net: Multipath Excitation for Action Recognition (246 citations)
- Temporal Context Aggregation Network for Temporal Action Proposal Refinement (183 citations)
- BABEL: Bodies, Action and Behavior With English Labels (182 citations)
- CoLA: Weakly-Supervised Temporal Action Localization With Snippet Contrastive Learning (166 citations)
- Actor-Context-Actor Relation Network for Spatio-Temporal Action Localization (144 citations)
- Deep Analysis of CNN-Based Spatio-Temporal Representations for Action Recognition (111 citations)
- Uncertainty Guided Collaborative Training for Weakly Supervised Temporal Action Detection (109 citations)
- Semi-Supervised Action Recognition With Temporal Contrastive Learning (100 citations)
and 38 more in the interactive map.
Video-language understanding · 21 papers
- Less Is More: ClipBERT for Video-and-Language Learning via Sparse Sampling (577 citations)
- NExT-QA: Next Phase of Question-Answering to Explaining Temporal Actions (284 citations)
- Interventional Video Grounding With Dual Contrastive Learning (151 citations)
- Context-Aware Biaffine Localizing Network for Temporal Sentence Grounding (141 citations)
- Towards Long-Form Video Understanding (120 citations)
- Open-Book Video Captioning With Retrieve-Copy-Generate Network (104 citations)
- Structured Multi-Level Interaction Network for Video Moment Localization via Language Query (100 citations)
- Temporal Query Networks for Fine-Grained Video Understanding (93 citations)
- AGQA: A Benchmark for Compositional Spatio-Temporal Reasoning (86 citations)
- Multi-Stage Aggregated Transformer Network for Temporal Language Localization in Videos (86 citations)
- Cascaded Prediction Network via Segment Tree for Temporal Video Grounding (85 citations)
- SUTD-TrafficQA: A Question Answering Benchmark and an Efficient Network for Video Reasoning Over Traffic Events (85 citations)
and 9 more in the interactive map.
Video generation · 6 papers
- One-Shot Free-View Neural Talking-Head Synthesis for Video Conferencing (445 citations)
- Flow-Guided One-Shot Talking Face Generation With a High-Resolution Audio-Visual Dataset (351 citations)
- Audio-Driven Emotional Video Portraits (206 citations)
- LipSync3D: Data-Efficient Learning of Personalized 3D Talking Faces From Video Using Pose and Lighting Normalization (100 citations)
- Playable Video Generation (36 citations)
- High-Fidelity Neural Human Motion Transfer From Monocular Video (33 citations)
