search

microsoft / View-Adaptive-Neural-Networks-for-Skeleton-based-Human-Action-Recognition

View Adaptive Neural Networks for High Performance Skeleton-based Human Action Recognition
MIT License
127 stars 58 forks source link

Will this model work on jpg image. Or we even need to pass depth information along with image during inference ? #10

Open bokey007 opened 4 years ago

bokey007 commented 4 years ago

I see this model was trained on NTU60 RGB+D. Was depth information used during training?

Will this model work on plin rgb images during inference or depht info in needed.

Also can I directly pass the pose estimations given bt openpose or alphapose or densepose ?