This research paper was published at ACM SIGGRAPH 2021.
Direct Delta Mush (DDM) is a high-quality, direct skinning method with a low setup cost. However, its storage and run-time computing cost are relatively high. This is for two reasons: its skinning weights are 4×4 matrices instead of scalars like other direct skinning methods, and its computation requires one 3×3 Singular Value Decomposition per vertex.
In this research paper, Binh Huy Le (SEED), Keven Villeneuve (SEED) and Carlos Gonzalez-Ochoa introduce a compression method into DDM to address the storage and runtime overhead.
This approach takes a DDM model and splits it into two layers: the first layer is a smaller DDM model that computes a set of virtual bone transformations, and the second layer is a Linear Blend Skinning model that computes per-vertex transformations from the output of the first layer. This two-layer model can approximate the deformation of the original DDM model with significantly lower costs.
The team’s main contribution is a novel problem formulation for the DDM compression based on a continuous example-based technique, in which they minimize the compression error on an uncountable set of example poses. This formulation provides an elegant metric for the compression error and simplifies the problem to the common linear matrix factorization. Their formulation also takes into account the skeleton hierarchy of the model, the bind pose, and the range of motions. In addition, they propose a new update rule to optimize DDM weights of the first layer and a modification to resolve the floating-point cancellation issue of DDM.
Watch the demonstration video above.
Download the reseatrch paper “DDM Skinning Compression with Continuous Examples” (PDF 25 MB)
Watch the full presentation below.