Support OpenPose tflite model (multi-pose)

[tucan9389]

Post-process Logic

• 1. Get points from confidence map (make nodes) [PR] Implement non maximum suppression for multi pose estimation #38
• 2. Make a bipartite graph with each part of points (make edges) [PR] Implement building bipartite graph, calculate cost values of each edge, assign and make humans #42
• 3. Calculate line integral with PAF (compute costs of each edge) [PR] Implement building bipartite graph, calculate cost values of each edge, assign and make humans #42
• 4. Connect each point and make human (solve assignment) [PR] Implement building bipartite graph, calculate cost values of each edge, assign and make humans #42
• 5. Merge if a point is used duplicated in multiple human [PR] Implement building bipartite graph, calculate cost values of each edge, assign and make humans #42

Checklist

• Add parameter control UIs for PAF Add post-processing param control UIs #47
• Limit the number of human on post-process Limit maximum number of human in multi-pose post process step #48
• Test other model

Reference

• https://arvrjourney.com/human-pose-estimation-using-openpose-with-tensorflow-part-2-e78ab9104fc8
• OpenPose: Realtime Multi-Person 2D Pose Estimation using Part Affinity Fields
• https://github.com/ildoonet/tf-pose-estimation