AAAI 2023 accepted papers on Video
81 of the 2,023 papers accepted at AAAI 2023 (4.0%) are on Video, compared with 3.9% at AAAI 2022. They fall into 4 topics.
See AAAI 2023 in the interactive map
Working on something in this area? Describe your idea in scime atlas to see the closest accepted papers by meaning and how crowded the spot has become.
Topics and papers
Video segmentation and tracking · 39 papers
- MGFN: Magnitude-Contrastive Glance-and-Focus Network for Weakly-Supervised Video Anomaly Detection (189 citations)
- Dual Memory Units with Uncertainty Regulation for Weakly Supervised Video Anomaly Detection (165 citations)
- Noise Based Deepfake Detection via Multi-Head Relative-Interaction (91 citations)
- Self-Emphasizing Network for Continuous Sign Language Recognition (80 citations)
- Intensity-Aware Loss for Dynamic Facial Expression Recognition in the Wild (73 citations)
- TrEP: Transformer-Based Evidential Prediction for Pedestrian Intention with Uncertainty (68 citations)
- DINet: Deformation Inpainting Network for Realistic Face Visually Dubbing on High Resolution Video (61 citations)
- Multi-Stream Representation Learning for Pedestrian Trajectory Prediction (61 citations)
- Trafformer: Unify Time and Space in Traffic Prediction (55 citations)
- Learning Motion-Robust Remote Photoplethysmography through Arbitrary Resolution Videos (52 citations)
- MIMO Is All You Need：A Strong Multi-in-Multi-Out Baseline for Video Prediction (44 citations)
- Low-Light Video Enhancement with Synthetic Event Guidance (36 citations)
and 27 more in the interactive map.
Action recognition · 20 papers
- Self-Supervised Action Representation Learning from Partial Spatio-Temporal Skeleton Sequences (54 citations)
- Revisiting the Spatial and Temporal Modeling for Few-Shot Action Recognition (50 citations)
- Novel Motion Patterns Matter for Practical Skeleton-Based Action Recognition (39 citations)
- MultiAct: Long-Term 3D Human Motion Generation from Multiple Action Labels (32 citations)
- Anonymization for Skeleton Action Recognition (21 citations)
- Deepfake Video Detection via Facial Action Dependencies Estimation (20 citations)
- Audio-Visual Contrastive Learning with Temporal Self-Supervision (18 citations)
- Actionness Inconsistency-Guided Contrastive Learning for Weakly-Supervised Temporal Action Localization (14 citations)
- Self-Supervised Video Representation Learning via Latent Time Navigation (10 citations)
- Frame-Level Label Refinement for Skeleton-Based Weakly-Supervised Action Recognition (9 citations)
- Video-Audio Domain Generalization via Confounder Disentanglement (8 citations)
- MEID: Mixture-of-Experts with Internal Distillation for Long-Tailed Video Recognition (7 citations)
and 8 more in the interactive map.
Video-language understanding · 16 papers
- FakeSV: A Multimodal Benchmark with Rich Social Context for Fake News Detection on Short Video Platforms (103 citations)
- Refined Semantic Enhancement towards Frequency Diffusion for Video Captioning (55 citations)
- VLTinT: Visual-Linguistic Transformer-in-Transformer for Coherent Video Paragraph Captioning (40 citations)
- Hypotheses Tree Building for One-Shot Temporal Sentence Localization (21 citations)
- Curriculum Multi-Negative Augmentation for Debiased Video Grounding (20 citations)
- Phrase-Level Temporal Relationship Mining for Temporal Sentence Localization (17 citations)
- AVCAffe: A Large Scale Audio-Visual Dataset of Cognitive Load and Affect for Remote Work (15 citations)
- Leveraging Weighted Cross-Graph Attention for Visual and Semantic Enhanced Video Captioning Network (15 citations)
- VideoDubber: Machine Translation with Speech-Aware Length Control for Video Dubbing (15 citations)
- Video-Text Pre-training with Learned Regions for Retrieval (11 citations)
- Efficient End-to-End Video Question Answering with Pyramidal Multimodal Transformer (9 citations)
- Knowledge-Constrained Answer Generation for Open-Ended Video Question Answering (8 citations)
and 4 more in the interactive map.
Video generation · 6 papers
- FLAME: Free-Form Language-Based Motion Synthesis & Editing (139 citations)
- StyleTalk: One-Shot Talking Head Generation with Controllable Speaking Styles (75 citations)
- VIDM: Video Implicit Diffusion Models (53 citations)
- Human Joint Kinematics Diffusion-Refinement for Stochastic Motion Prediction (51 citations)
- Deep Visual Forced Alignment: Learning to Align Transcription with Talking Face Video (3 citations)
- Learning Adaptive Game Soundtrack Control