Predicting Animation Skeletons for 3D Articulated Models via Volumetric Nets
This addresses the need for automated skeleton generation in animation, offering a domain-specific improvement over template-based methods.
The paper tackles the problem of predicting animation skeletons for 3D articulated models by introducing a learning method that produces tailored skeletons based on model structure and geometry, resulting in skeletons much more similar to human-created ones compared to alternatives.
We present a learning method for predicting animation skeletons for input 3D models of articulated characters. In contrast to previous approaches that fit pre-defined skeleton templates or predict fixed sets of joints, our method produces an animation skeleton tailored for the structure and geometry of the input 3D model. Our architecture is based on a stack of hourglass modules trained on a large dataset of 3D rigged characters mined from the web. It operates on the volumetric representation of the input 3D shapes augmented with geometric shape features that provide additional cues for joint and bone locations. Our method also enables intuitive user control of the level-of-detail for the output skeleton. Our evaluation demonstrates that our approach predicts animation skeletons that are much more similar to the ones created by humans compared to several alternatives and baselines.