mediapipe/mediapipe/modules/pose_landmark/README.md
MediaPipe Team 2b58cceec9 Project import generated by Copybara.
GitOrigin-RevId: d8caa66de45839696f5bd0786ad3bfbcb9cff632
2020-12-09 22:43:33 -05:00

11 lines
2.7 KiB
Markdown

# pose_landmark
Subgraphs|Details
:--- | :---
[`PoseLandmarkByRoiCpu`](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_by_roi_cpu.pbtxt)| Detects landmarks of a single body pose, full-body by default but can be configured (via an input side packet) to cover upper-body only. See landmarks (aka keypoints) [scheme](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_full_body_topology.svg). (CPU input, and inference is executed on CPU.)
[`PoseLandmarkByRoiGpu`](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_by_roi_gpu.pbtxt)| Detects landmarks of a single body pose, full-body by default but can be configured (via an input side packet) to cover upper-body only. See landmarks (aka keypoints) [scheme](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_full_body_topology.svg). (GPU input, and inference is executed on GPU)
[`PoseLandmarkCpu`](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_cpu.pbtxt)| Detects landmarks of a single body pose, full-body by default but can be configured (via an input side packet) to cover upper-body only. See landmarks (aka keypoints) [scheme](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_full_body_topology.svg). (CPU input, and inference is executed on CPU)
[`PoseLandmarkGpu`](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_gpu.pbtxt)| Detects landmarks of a single body pose, full-body by default but can be configured (via an input side packet) to cover upper-body only. See landmarks (aka keypoints) [scheme](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_full_body_topology.svg). (GPU input, and inference is executed on GPU.)
[`PoseLandmarkUpperBodyCpu`](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_upper_body_cpu.pbtxt)| Detects and tracks landmarks of a single upper-body pose. See landmarks (aka keypoints) [scheme](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_upper_body_topology.svg). (CPU input, and inference is executed on CPU)
[`PoseLandmarkUpperBodyGpu`](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_upper_body_gpu.pbtxt)| Detects and tracks landmarks of a single upper-body pose. See landmarks (aka keypoints) [scheme](https://github.com/google/mediapipe/tree/master/mediapipe/modules/pose_landmark/pose_landmark_upper_body_topology.svg). (GPU input, and inference is executed on GPU.)