Dual Geometric Graph Network (DG2N) Iterative Network for Deformable Shape Alignment
Dvir Ginzburg and Dan Raviv
We provide a novel new approach for aligning geometric models using a dual graph structure where local features are mapping probabilities. The alignment of non-rigid structures is one of the most challenging computer vision tasks due to the high number of unknowns needed to model the correspondence. We have seen a leap forward using DNN models in template alignment and functional maps, but those methods fail for inter-class alignment where non-isometric deformations exist. Here we propose to rethink this task and use unrolling concepts on a dual graph structure - one for a forward map and one for a backward map, where the features are pulled back matching probabilities from the target into the source. We report state-of-the-art results on stretchable domains' alignment in a rapid and stable solution for meshes and clouds of points.