Part Segmentation Size in fit_up.py and demo.lua

gulvarol / bodynet

BodyNet: Volumetric Inference of 3D Human Body Shapes, ECCV 2018

http://www.di.ens.fr/willow/research/bodynet

MIT License

261 stars 42 forks source link

Part Segmentation Size in fit_up.py and demo.lua #16

Open liren2515 opened 4 years ago

liren2515 commented 4 years ago

Hello,

When I was trying to use the results from demo.lua to do the fitting implemented in fit_up.py, I found that the size of the 2D segmentation from the network is 256x256, but in fit_up.py, the size of the segmentation is 128x128. what should I do to resolve this size inconsistency? Like a downsampling or something?

liren2515 commented 4 years ago

Another question is how to convert the 3D joints inferred from the model to the coordinate used by fit_up.py.