We introduce Breaking Bad, a large-scale dataset of fractured objects.

To this end, we et al., 2023b) to address the task of shape assembly, but these introduce Proxy Match Transform (PMT), an methods fall short of achieving accurate assembly. They approximate high-order feature transform layer typically represent each part as a global embedding and that enables reliable matching between mating perform regression to predict a placement for each part. The surfaces of parts while incurring low costs in global encoding strategy for each part, while simplifying memory and computation. Building upon PMT, the process, greatly limits local information by collapsing we introduce a new framework, dubbed Proxy spatial resolutions, which is necessary to localize the mating Match TransformeR (PMTR), for the geometric surface. Indeed, accurate shape assembly requires a detailed assembly task. We evaluate the proposed PMTR analysis of both fine-and coarse-level spatial information on the large-scale 3D geometric shape assembly of the parts in recognizing mating surfaces and establishing benchmark dataset of Breaking Bad and demonstrate correspondences between the surfaces. Therefore, a its superior performance and efficiency compared promising approach would be to retain the spatially rich to state-of-the-art methods. Project page: part representations during the encoding phase and analyze https://nahyuklee.github.io/pmtr.

Shape assembly aims to reassemble parts (or fragments) into a complete object, which is a common task in our daily life. Different from the semantic part assembly (e.g., assembling a chair's semantic parts like legs into a whole chair), geometric part assembly (e.g., assembling bowl fragments into a complete bowl) is an emerging task in computer vision and robotics. Instead of semantic information, this task focuses on geometric information of parts. As the both geometric and pose space of fractured parts are exceptionally large, shape pose disentanglement of part representations is beneficial to geometric shape assembly. In our paper, we propose to leverage SE(3) equivariance for such shape pose disentanglement. Moreover, while previous works in vision and robotics only consider SE(3) equivariance for the representations of single objects, we move a step forward and propose leveraging SE(3) equivariance for representations considering multi-part correlations, which further boosts the performance of the multi-part assembly. Experiments demonstrate the significance of SE(3) equivariance and our proposed method for geometric shape assembly. Project page: https://crtie.github.io/SE-3-part-assembly/