Name That Part: 3D Part Segmentation and Naming

Soumava Paul, Prakhar Kaushik, Ankit Vaidya, Anand Bhattad, Alan Yuille

Keywords: 3D parts

Abstract: We address semantic 3D part segmentation: decomposing objects into parts with meaningful names. While datasets exist with part annotations, their definitions are inconsistent across datasets, limiting robust training. Previous methods produce unlabeled decompositions or retrieve single parts without complete shape annotations. We propose ALIGN-Parts, which formulates part naming as a direct set alignment task. Our method decomposes shapes into partlets - implicit 3D part representations - matched to part descriptions via bipartite assignment. We combine geometric cues from 3D part fields, appearance from multi-view vision features, and semantic knowledge from language-model-generated affordance descriptions. Text-alignment loss ensures partlets share embedding space with text, enabling open-vocabulary matching. Our efficient and novel, fast one-shot, 3D part segmentation and naming method finds usage in several downstream tasks, including as a scalable annotation engine. As our model supports zero-shot matching to arbitrary descriptions and confidence-calibrated predictions for known categories - with human verification - we create a unified ontology aligning PartNet, 3DCoMPaT++, and Find3D, consisting of 1794 unique 3D parts, and show examples from our newly created Tex-Parts dataset.

Previously Accepted: Yes

Previous Venue: CVPR 2026 Findings

Email Sharing: We authorize the sharing of all author emails with Program Chairs.

Data Release: We authorize the release of our submission and author names to the public in the event of acceptance.

Submission Number: 5