Skip to content

Solving SMPL/MANO parameters from keypoint coordinates.

License

Notifications You must be signed in to change notification settings

CalciferZh/Minimal-IK

Repository files navigation

Minimal-IK

A simple and naive inverse kinematics solver for MANO hand model, SMPL body model, and SMPL-H body+hand model.

Briefly, given joint coordinates (and optional other keypoints), the solver gives the corresponding model parameters.

Levenberg–Marquardt algorithm is used, the energy is simply the L2 distance between the keypoints.

As no prior nor regularization terms are used, it is not surprising that the code does not work well on "real" data. My intention to release the code was to give some hints on how to develope a customized IK solver. I would recommend to add more complicating terms for better performance.

Results

Qualitative

This is the example result on the SMPL body model. The left is the ground truth, and the right one is the estimation. You can notice the minor difference between the left hands.

Below is the example result of the MANO hand model. Left for ground truth, and right for estimation.

Quantitative

We test this approach on the AMASS dataset.

Mean Joint Error (mm) Mean Vertex Error (mm)
SMPL (body) 14.406 23.110
MANO (hand) 2.15 3.42

We assume that the global rotation is known. We discuss this further in the Customization Notes section.

Usage

Models

  1. Download the official model from MPI.
  2. See config.py and set the official model path.
  3. See prepare_model.py, use the provided function to pre-process the model.

Solver

  1. See example.py, un-comment the corresponding code.
  2. python example.py.
  3. The example ground truth mesh and estimated mesh are saved to gt.obj and est.obj respectively.

Dependencies

Every required package is available via pip install.

Customization Notes

Again, we note that this approach cannot handle large global rotations (R0) due to the high non-convexity. For example, when the subject keeps the T pose but faces backwards.

In such cases, a good initialization, at least for R0, is necessary.

We also note that this approach is sensitive the the scale (i.e. length unit), as it would affect the MSE and the update step. Please consider using the default scale if you do not have special reasons.

Credits

  • @yxyyyxxyy for the quantitative test on the AMASS dataset.
  • @zjykljf for the starter code of the LM solver.

About

Solving SMPL/MANO parameters from keypoint coordinates.

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages