TOSAI project

Understanding and Creating Dynamic 3D Worlds towards Safer AI

Code repositories
ControlNet-XS	https://github.com/vislearn/ControlNet-XS
H3WB: Wholebody 3D pose estimation	https://github.com/wholebody3d/wholebody3d
DeePoint: Visual Pointing Recognition and Direction Estimation	https://github.com/kyotovision-public/deepoint
nLMVS-Net: Deep Non-Lambertian Multi-View Stereo	https://github.com/kyotovision-public/nLMVS-Net
Decanus synthetic training for pose lifting	https://github.com/Zhuyue0324/Decanus-to-Legatus
SCAM: semantic cross-attention modulation	https://github.com/nicolas-dufour/SCAM
Neural Head Avatars	https://github.com/philgras/neural-head-avatars
Multimodal Material Segmentation	https://github.com/kyotovision-public/multimodal-material-segmentation
Dynamic 3D Gaze from Afar	https://github.com/kyotovision-public/dynamic-3d-gaze-from-afar
Shape from Sky: Polarimetric Normal Recovery Under the Sky	https://github.com/kyotovision-public/shape-from-sky
Polarimetric Normal Stereo Dataset	https://github.com/kyotovision-public/polarimetric-normal-stereo
Trustworthy generative classifiers	https://github.com/vislearn/trustworthy_GCs

Code repositories