AAAI 2022 accepted papers on Video
52 of the 1,319 papers accepted at AAAI 2022 (3.9%) are on Video, compared with 3.3% at AAAI 2021. They fall into 4 topics.
See AAAI 2022 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 · 31 papers
- Self-Training Multi-Sequence Learning with Transformer for Weakly Supervised Video Anomaly Detection (279 citations)
- Delving into the Local: Dynamic Inconsistency Learning for DeepFake Video Detection (89 citations)
- FInfer: Frame Inference-Based Deepfake Detection for High-Visual-Quality Videos (85 citations)
- Comprehensive Regularization in a Bi-directional Predictive Network for Video Anomaly Detection (76 citations)
- MobileFaceSwap: A Lightweight Framework for Video Face Swapping (67 citations)
- Spatio-Temporal Recurrent Networks for Event-Based Optical Flow Estimation (65 citations)
- Learning Optical Flow with Adaptive Graph Reasoning (62 citations)
- Complementary Attention Gated Network for Pedestrian Trajectory Prediction (54 citations)
- Deep Recurrent Neural Network with Multi-Scale Bi-directional Propagation for Video Deblurring (53 citations)
- Reliable Propagation-Correction Modulation for Video Object Segmentation (53 citations)
- Self-Supervised Representation Learning Framework for Remote Physiological Measurement Using Spatiotemporal Augmentation Loss (52 citations)
- A Causal Inference Look at Unsupervised Video Anomaly Detection (42 citations)
and 19 more in the interactive map.
Action recognition · 11 papers
- TA2N: Two-Stage Action Alignment Network for Few-Shot Action Recognition (87 citations)
- Towards To-a-T Spatio-Temporal Focus for Skeleton-Based Action Recognition (73 citations)
- DCAN: Improving Temporal Action Detection via Dual Context Aggregation (69 citations)
- ACGNet: Action Complement Graph Network for Weakly-Supervised Temporal Action Localization (67 citations)
- Iterative Contrast-Classify for Semi-supervised Temporal Action Segmentation (26 citations)
- Temporal Action Proposal Generation with Background Constraint (23 citations)
- Self-Supervised Spatiotemporal Representation Learning by Exploiting Video Continuity (20 citations)
- Learning Disentangled Classification and Localization Representations for Temporal Action Localization (15 citations)
- Shrinking Temporal Attention in Transformers for Video Action Recognition (14 citations)
- Learning from Weakly-Labeled Web Videos via Exploring Sub-concepts (7 citations)
- Towards Explainable Action Recognition by Salient Qualitative Spatial Object Relation Chains (6 citations)
Video-language understanding · 6 papers
- Negative Sample Matters: A Renaissance of Metric Learning for Temporal Grounding (130 citations)
- Weakly Supervised Video Moment Localization with Contrastive Negative Sample Mining (104 citations)
- Memory-Guided Semantic Learning Network for Temporal Sentence Grounding (70 citations)
- Unsupervised Temporal Video Grounding with Deep Semantic Clustering (43 citations)
- Explore Inter-contrast between Videos via Composition for Weakly Supervised Temporal Sentence Grounding (26 citations)
- Visual Consensus Modeling for Video-Text Retrieval (19 citations)
Video generation · 4 papers
- One-Shot Talking Face Generation from Single-Speaker Audio-Visual Correlation Learning (109 citations)
- SyncTalkFace: Talking Face Generation with Precise Lip-Syncing via Audio-Lip Memory (87 citations)
- Flow-Based Unconstrained Lip to Speech Generation (19 citations)
- Parallel and High-Fidelity Text-to-Lip Generation (7 citations)
