← Back to Model Beat
4Research·5h ago

Multi-THuMBS: Multi-person Tracking of 3D Human Meshes Beyond Video Shots

Researchers have introduced a new framework called Multi-THuMBS designed to track three-dimensional human body meshes for multiple people in unconstrained video environments. The method aims to overcome common technical hurdles such as subjects moving behind objects, overlapping each other, or partially disappearing from the frame. By improving the consistency of these reconstructions across varying camera angles and complex settings, the approach provides a more reliable way to interpret human motion in challenging real-world recordings.

Covered by 1 source

  • AarXiv CS.AIJeongwan On, Muhammad Salman Ali, Muneeb A. Khan, Sunwoo Park, Inwoong Moon, Hyung Jin Chang, Jaekwang Kim, Seong Jong Ha, Seungryul Baek5h ago

Related stories

ResearchOn Robustness and Chain-of-Thought Consistency of RL-Finetuned VLMsJun 29 · 13 sourcesResearchAnti-Causal Domain Generalization: Leveraging Unlabeled DataJul 1 · 2 sourcesResearchLearning Unmasking Policies for Diffusion Language ModelsJun 29 · 6 sourcesResearchRedKnot: Efficient Long-Context LLM Serving with Head-Aware KV Reuse and SegPagedAttentionJun 29