video understanding
Video understanding refers to the AI capability to analyze and interpret video content by recognizing actions, events, and contextual elements across time. This area of research intertwines computer vision and natural language processing to enable applications such as surveillance and content analysis.
- AdaVideoRAG: Omni-Contextual Adaptive Retrieval-Augmented Efficient Long Video Understanding
- Bisecle: Binding and Separation in Continual Learning for Video Language Understanding
- EgoExoBench: A Benchmark for First- and Third-person View Video Understanding in MLLMs
- Enhancing Temporal Understanding in Video-LLMs through Stacked Temporal Attention in Vision Encoders
- Mitigating Hallucination in VideoLLMs via Temporal-Aware Activation Engineering
- One Token per Highly Selective Frame: Towards Extreme Compression for Long Video Understanding
- PerceptionLM: Open-Access Data and Models for Detailed Visual Understanding
- Prompt-guided Disentangled Representation for Action Recognition
- ReAgent-V: A Reward-Driven Multi-Agent Framework for Video Understanding
- SAGE: A Unified Framework for Generalizable Object State Recognition with State-Action Graph Embedding
- SAGE: A Unified Framework for Generalizable Object State Recognition with State-Action Graph Embedding
- StreamForest: Efficient Online Video Understanding with Persistent Event Memory
- Time-R1: Post-Training Large Vision Language Model for Temporal Video Grounding
- VideoCAD: A Dataset and Model for Learning Long‑Horizon 3D CAD UI Interactions from Video
- VideoHallu: Evaluating and Mitigating Multi-modal Hallucinations on Synthetic Video Understanding