Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image
2017-01-01CVPR 2017Code Available0· sign in to hype
Denis Tome, Chris Russell, Lourdes Agapito
Code Available — Be the first to reproduce this paper.
ReproduceCode
- github.com/SyBorg91/pose-estimation-detectiontf★ 0
- github.com/DenisTome/Lifting-from-the-Deep-releasetf★ 0
- github.com/h44rd/PoseTransferMayaPluginnone★ 0
- github.com/anfidthtn/HACtf★ 0
- github.com/SyedUmaidAhmed/Pose-Detection-Sitting-Standingtf★ 0
- github.com/Watson-BCA/HumanPostureRecognitiontf★ 0
- github.com/zhec/tf-pose-estimationtf★ 0
- github.com/satyaborg/pose-estimation-detectiontf★ 0
- github.com/swathi-469/Teresa-Bottf★ 0
- github.com/sourabhagrawal23/human-action-classificationtf★ 0
Abstract
We propose a unified formulation for the problem of 3D human pose estimation from a single raw RGB image that reasons jointly about 2D joint estimation and 3D pose reconstruction to improve both tasks. We take an integrated approach that fuses probabilistic knowledge of 3D human pose with a multi-stage CNN architecture and uses the knowledge of plausible 3D landmark locations to refine the search for better 2D locations. The entire process is trained end-to-end, is extremely efficient and obtains state- of-the-art results on Human3.6M outperforming previous approaches both on 2D and 3D errors.