[Open-source Project] UniMoCap: community implementation to unify the text-motion datasets (HumanML3D, KIT-ML, and BABEL) and whole-body motion dataset (Motion-X).

🎭 Enhance facial expression recognition using multimodal large language models to improve emotion analysis in various applications.

Repo for portfolio, containing working redirects to all projects.

Text to motion model (pytorch) trained on HumanML3D dataset

Fork of MotionGPT3 adapted for synthetic rehabilitation-motion generation (the RehabMotion pipeline): adds multi-task fine-tuning configs, rehab caption/instruction templates, and HumanML3D data hooks for the KIMORE and UI-PRMD exercise datasets.