PatchComplete: Learning Multi-Resolution Patch Priors for 3D Shape Completion on Unseen Categories

NeurIPS 2022 Paper Video
Paper: arxiv.org/abs/...
Project webpage: yuchenrao.gith...
While 3D shape representations enable powerful reasoning in many visual and perception applications, learning 3D shape priors tends to be constrained to the specific categories trained on, leading to an inefficient learning process, particularly for general applications with unseen categories. Thus, we propose PatchComplete, which learns effective shape priors based on multi-resolution local patches, which are often more general than full shapes (e.g., chairs and tables often both share legs) and thus enable geometric reasoning about unseen class categories. To learn these shared substructures, we learn multi-resolution patch priors across all train categories, which are then associated to input partial shape observations by attention across the patch priors, and finally decoded into a complete shape reconstruction. Such patch-based priors avoid overfitting to specific train categories and enable reconstruction on entirely unseen categories at test time. We demonstrate the effectiveness of our approach on synthetic ShapeNet data as well as challenging real-scanned objects from ScanNet, which include noise and clutter, improving over state of the art in novel-category shape completion by 19.3% in chamfer distance on ShapeNet, and 9.0% for ScanNet.

[@NegarNejatishahidyan]

I am wondering if the initial scans represented in object coordinate space? (specially for the real-data) I am interested to understand if you have used ground-truth poses to bring them into object coordinate space.