Has HybrIK been tested against the SoTA 3D pose and shape estimation methods that work on the SMPLX body model (Ex: PIXIE, SMPLify-X)?
Since compared to the SMPL body model, SMPLX also allows for movement of the finger joints, and it would be interesting to see how the Hybrid Analytical-Neural approach of HybrIK compares against the Regression/Optimization based approaches of the SoTA methods.
Has HybrIK been tested against the SoTA 3D pose and shape estimation methods that work on the SMPLX body model (Ex: PIXIE, SMPLify-X)? Since compared to the SMPL body model, SMPLX also allows for movement of the finger joints, and it would be interesting to see how the Hybrid Analytical-Neural approach of HybrIK compares against the Regression/Optimization based approaches of the SoTA methods.