EvObj: Learning Evolving Object-centric Representations for 3D Instance Segmentation without Scene Supervision

TL;DR

EvObj is an unsupervised 3D instance segmentation method that adapts object priors from synthetic to real data, using dynamic candidate refinement and geometry reconstruction.

Contribution

It introduces two modules for domain adaptation: dynamic object candidate refinement and partial geometry reconstruction, improving real-world segmentation.

Findings

Outperforms all baseline methods on real-world and synthetic datasets.

Achieves state-of-the-art results in 3D object segmentation.

Effectively bridges the gap between synthetic pretraining data and real scans.

Abstract

We introduce EvObj for unsupervised 3D instance segmentation that bridges the geometric domain gap between synthetic pretraining data and real-world point clouds. Current methods suffer from structural discrepancies when transferring object priors from synthetic datasets (e.g., ShapeNet) to real scans (e.g., ScanNet), particularly due to morphological variations and occlusion artifacts. To address this, EvObj integrates two innovative modules: (1) An object discerning module that dynamically refines object candidates, enabling continuous adaptation of object priors to target domains; and (2) An object completion module that reconstructs partial geometries after discovering objects. We conduct extensive experiments on both real-world and synthetic datasets, demonstrating superior 3D object segmentation performance over all baselines while achieving state-of-the-art results.