This is a C++ version of VNect[1] demo. And this demo refers to the matlab demo[1][2] of the authors.
The fitting process and the kenimatic skeleton is not like the one in the paper(I will learn about the inverse kenimatic and rigging latter. T T).
- VNect: Real-time 3D Human Pose Estimation with a Single RGB Camera Mehta, D.; Sridhar, S.; Sotnychenko, O.; Rhodin, H.; Shafiei, M.; Seidel, H.; Xu, W.; Casas, D.; Theobalt, C. ACM Transactions on Graphics, Proceedings SIGGRAPH 2017
- Monocular 3D Human Pose Estimation In The Wild Using Improved CNN Supervision Mehta, D.; Rhodin, H.; Casas, D.; Sotnychenko, O.; Xu, W.; Theobalt, C. arxiv:1611.09813
- Model-based Outdoor Performance Capture Robertini, N.; Casas, D.; Rhodin, H.; Seidel, H.P.S.; Theobalt, C. Proceedings of the Second International Conference on 3D Vision (3DV), 2016
- The long computation time in the fitting process may lie in the derivatives.