Pinned Loading

1. [NeurIPS 2022 Spotlight] VideoMAE: Masked Autoencoders are Data-Efficient Learners for Self-Supervised Video Pre-Training

   Python 1.8k 170

2. A generalist video MLLM built for fine-grained motion, long-form reasoning, temporal grounding, and online proactive response.

   Python 397 8

3. [CVPR 2022 Oral & TPAMI 2024] MixFormer: End-to-End Tracking with Iterative Mixed Attention

   Python 540 75

4. [CVPR 2025] Multiple Object Tracking as ID Prediction

   Python 553 53

5. [NeurIPS 2025 Spotlight] StreamForest: Efficient Online Video Understanding with Persistent Event Memory

   Python 129 5

6. [CVPR 2026] DDT: Decoupled Diffusion Transformer

   Python 416 22