Multi-person 3D human pose estimation model based on the Lightweight OpenPose and Single-Shot Multi-Person 3D Pose Estimation From Monocular RGB papers.
| Metric | Value |
|---|---|
| MPJPE (mm) | 100.45 |
| GFlops | 18.998 |
| MParams | 5.074 |
| Source framework | PyTorch* |
Name: data, shape: [1x3x256x448]. An input image in the [BxCxHxW] format,
where:
- B - batch size
- C - number of channels
- H - image height
- W - image width
Expected color order is BGR.
The net outputs three blobs with the following shapes: [1, 57, 32, 56], [1, 19, 32, 56], and [1, 38, 32, 56]. The first blob contains coordinates in 3D space, the second blob contains keypoint heatmaps and the third blob is keypoint pairwise relations (part affinity fields).
The original model is distributed under the Apache License, Version 2.0. A copy of the license is provided in APACHE-2.0-TensorFlow.txt.
[*] Other names and brands may be claimed as the property of others.