search

suyz526 / ZebraPose

The implementation of the paper 'ZebraPose: Coarse to Fine Surface Encoding for 6DoF Object Pose Estimation' (CVPR2022)
MIT License
122 stars 20 forks source link

loss for predicted binary coding #36

Closed hrkita closed 6 months ago

hrkita commented 10 months ago

Hi，

If I want to compare the entire mask to obtain a loss function, what part of the code do I need to modify? Or do I need to retrain the ResNet34 model?

if predict_entire_mask: loss_m = maskLoss(pred_mask_prob, entire_masks) else: loss_m = maskLoss(pred_mask_prob, masks)

Best, Joa

suyz526 commented 9 months ago

If you set this arg as True, the network will predict entire mask instead of visible mask.

hrkita commented 9 months ago

Okay, I will give it a try.Thanks!