Every Pixel Counts : Joint Learning of Geometry and Motion with 3D Holistic Understanding This is a Tensorflow implementation of our paper: Every Pixel Counts : Joint Learning of Geometry and Motion with 3D Holistic Understanding (TPAMI accepted) Prerequisites This codebase was developed and tested with Tensorflow 1.8, CUDA 9.0 and Ubuntu 16.04. Pretrained models: link Results mono stereo