The authors of EffiScene: Efficient Per-Pixel Rigidity Inference for Unsupervised Joint Learning of Optical Flow, Depth, Camera Pose and Motion Segmentation have not publicly listed the code yet.

Request code directly from the authors: Ask Authors for Code

Get an expert to implement this paper: Request Implementation

(OR if you have code to share with the community, please add it here 😊🙏 )