OpenMaskDINO3D : Reasoning 3D Segmentation via Large Language Model
Kunshen Zhang · 2025
Read OpenMaskDINO3D : Reasoning 3D Segmentation via Large Language Model by Kunshen Zhang (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2506-04837.txt.
Description
Kunshen Zhang takes three-dimensional segmentation beyond fixed labels in a 2025 computer vision study. Two-dimensional systems can already cut out objects described by hints about intent, not just category names. Comparable reasoning in three dimensions, where a query might describe a part by its use or relation, has lacked a framework.
The proposed system couples a detection transformer design with a large language model that interprets the query. Masks in point cloud scenes then follow from indirect descriptions. Open reasoning segmentation would let robots and design tools respond to requests like the part you sit on, without a preset label list.
Catalog record
| CREATOR | Kunshen Zhang |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 28 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2506-04837