High-fidelity Interpretable Inverse Rig: An Accurate and Sparse Solution Optimizing the Quartic Blendshape Model

TL;DR

This paper introduces a novel inverse rig optimization method for human face animation that achieves highly accurate, sparse blendshape solutions with improved fit quality and easier manual manipulation, outperforming existing approaches.

Contribution

It presents a coordinate descent-based approach for inverse rig fitting that ensures mutual exclusivity of controllers and reduces weight vector cardinality, enhancing accuracy and interpretability.

Findings

Achieves mesh error comparable or lower than state-of-the-art methods.

Reduces weight vector cardinality by over 20%.

Produces high-fidelity, manipulable face reconstructions.

Abstract

We propose a method to fit arbitrarily accurate blendshape rig models by solving the inverse rig problem in realistic human face animation. The method considers blendshape models with different levels of added corrections and solves the regularized least-squares problem using coordinate descent, i.e., iteratively estimating blendshape weights. Besides making the optimization easier to solve, this approach ensures that mutually exclusive controllers will not be activated simultaneously and improves the goodness of fit after each iteration. We show experimentally that the proposed method yields solutions with mesh error comparable to or lower than the state-of-the-art approaches while significantly reducing the cardinality of the weight vector (over 20 percent), hence giving a high-fidelity reconstruction of the reference expression that is easier to manipulate in the post-production manually. Python scripts for the algorithm will be publicly available upon acceptance of the paper.