Physically Disentangled Representations

1 code implementation • 11 Apr 2022 • Tzofi Klinghoffer, Kushagra Tiwary, Arkadiusz Balata, Vivek Sharma, Ramesh Raskar

In this paper, we show the utility of inverse rendering in learning representations that yield improved accuracy on downstream clustering, linear classification, and segmentation tasks with the help of our novel Leave-One-Out, Cycle Contrastive loss (LOOCC), which improves disentanglement of scene parameters and robustness to out-of-distribution lighting and viewpoints.

Attribute Classification +3