Lifting from the Deep: Convolutional 3D Pose Estimation from a Single Image
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.
PDF Abstract CVPR 2017 PDF CVPR 2017 AbstractCode
Datasets
Human3.6M
Results from the Paper
| Task | Dataset | Model | Metric Name | Metric Value | Global Rank | Benchmark |
|---|---|---|---|---|---|---|
| Weakly-supervised 3D Human Pose Estimation | Human3.6M | Tome et al. | Average MPJPE (mm) | 88.4 | # 22 | |
| Number of Views | 1 | # 1 | ||||
| Number of Frames Per View | 1 | # 1 | ||||
| 3D Annotations | No | # 1 | ||||
| Monocular 3D Human Pose Estimation | Human3.6M | Projected-pose belief maps + 2D fusion layers | Average MPJPE (mm) | 88.39 | # 36 | |
| Use Video Sequence | No | # 1 | ||||
| Frames Needed | 1 | # 1 | ||||
| Need Ground Truth 2D Pose | No | # 1 |
Results from Other Papers
Methods
No methods listed for this paper. Add
relevant methods here
