Shoukang Hu
Scholar

Shoukang Hu

Google Scholar ID: 9cUPotAAAAAJ
Microsoft Research
3D Human AvatarAudio/Speech ProcessingAutomated Machine Learning
Citations & Impact
All-time
Citations
1,806
 
H-index
24
 
i10-index
41
 
Publications
20
 
Co-authors
17
list available
Resume (English only)
Academic Achievements
  • Paper 'Free4D: Tuning-free 4D Scene Generation with Spatial-Temporal Consistency' accepted by ICCV 2025
  • Paper 'HumanLiff: Layer-wise 3D Human Generation with Diffusion Model' accepted by IJCV
  • Paper 'WildAvatar: Web-scale In-the-wild Video Dataset for 3D Avatar Creation' accepted by CVPR 2025
  • Preprint 'HumanGif: Single-View Human Diffusion with Generative Prior' released
  • Paper 'MVSGaussian: Fast Generalizable Gaussian Splatting Reconstruction from Multi-View Stereo' accepted by ECCV 2024
  • Paper 'GauHuman: Articulated Gaussian Splatting from Monocular Human Videos' accepted by CVPR 2024
  • Co-organizing the 2nd Workshop on Audio-Visual Generation and Learning at ICCV 2025
  • Developed notable projects including HumanGif, GauHuman, HumanLiff, SHERF, ConsistentNeRF, GM-NAS, NAS-Analysis, DSNAS, etc.