Segment3D: Learning Fine-Grained Class-Agnostic 3D Segmentation Without Manual Labels


METADATA ONLY
Loading...

Date

2025

Publication Type

Conference Paper

ETH Bibliography

yes

Citations

Altmetric
METADATA ONLY

Data

Rights / License

Abstract

Current 3D scene segmentation methods are heavily dependent on manually annotated 3D training datasets. Such manual annotations are labor-intensive, and often lack fine-grained details. Furthermore, models trained on this data typically struggle to recognize object classes beyond the annotated training classes, i.e., they do not generalize well to unseen domains and require additional domain-specific annotations. In contrast, recent 2D foundation models have demonstrated strong generalization and impressive zero-shot abilities, inspiring us to incorporate these characteristics from 2D models into 3D models. Therefore, we explore the use of image segmentation foundation models to automatically generate high-quality training labels for 3D segmentation models. The resulting model, Segment3D, generalizes significantly better than the models trained on costly manual 3D labels and enables easily adding new training data to further boost the segmentation performance.

Publication status

published

Book title

Computer Vision – ECCV 2024

Volume

15092

Pages / Article No.

278 - 295

Publisher

Springer

Event

18th European Conference on Computer Vision (ECCV 2024)

Edition / version

Methods

Software

Geographic location

Date collected

Date created

Subject

Class-agnostic 3D segmentation; 3D Scene Understanding

Organisational unit

03766 - Pollefeys, Marc / Pollefeys, Marc check_circle
02154 - Media Technology Center (MTC) / Media Technology Center (MTC)

Notes

Funding

Related publications and datasets