Mingyuan Zhang
2 papers · Latest:
Computer Vision
MoCapAnything V2: End-to-End Motion Capture for Arbitrary Skeletons
MoCapAnything V2 is an end-to-end system for arbitrary-skeleton motion capture from monocular video, achieving higher accuracy and 20x faster inference.
2604.28130
Computer VisionDistorted or Fabricated? A Survey on Hallucination in Video LLMs
This survey categorizes and analyzes hallucinations in Video LLMs, detailing their types, causes, evaluation, and mitigation strategies.
2604.12944
📬 Weekly AI Paper Digest
Get the top 10 AI/ML arXiv papers from the week — summarized, scored, and delivered to your inbox every Monday.