NeurIPS 2026Spotlight
Improving the Diffusability of Motion Tokenizer
Post-training motion tokenizers with spectral latent regularization and generative decoding for diffusion-friendly latent spaces.
VIRTUAL HUMAN · GENERATIVE AI · HUMAN–ROBOT INTERACTION
Hi, I’m
Guanhe Huang.
I am pursuing a PhD in Engineering Research at King’s College London, supervised by Dr Oya Celiktutan. My research focuses on deep generative human motion modeling for human–robot interaction.
Selected Publications
All publications →
CVPR 2026Highlight
Unified Number-Free Text-to-Motion Generation via Flow Matching
Unified flow matching for variable-person motion generation, reducing computation and stage-wise error accumulation.