The repository that this is forked from is an official PyTorch implementation of the paper "Learnable Triangulation of Human Pose" (ICCV 2019, oral). Proposed method archives state-of-the-art results in multi-view 3D human pose estimation! This repo aims to extend the capabilites to the CMU dataset
Where is the file mrcnn-detections?