Part2Object: Hierarchical Unsupervised 3D Instance Segmentation
This addresses the challenge of segmenting objects from 3D point clouds without annotations, which is incremental as it builds on prior clustering methods.
The paper tackles the problem of unsupervised 3D instance segmentation, where existing methods suffer from under- or over-segmentation, by proposing Part2Object with hierarchical clustering and objectness priors, achieving superior performance compared to state-of-the-art models in various settings.
Unsupervised 3D instance segmentation aims to segment objects from a 3D point cloud without any annotations. Existing methods face the challenge of either too loose or too tight clustering, leading to under-segmentation or over-segmentation. To address this issue, we propose Part2Object, hierarchical clustering with object guidance. Part2Object employs multi-layer clustering from points to object parts and objects, allowing objects to manifest at any layer. Additionally, it extracts and utilizes 3D objectness priors from temporally consecutive 2D RGB frames to guide the clustering process. Moreover, we propose Hi-Mask3D to support hierarchical 3D object part and instance segmentation. By training Hi-Mask3D on the objects and object parts extracted from Part2Object, we achieve consistent and superior performance compared to state-of-the-art models in various settings, including unsupervised instance segmentation, data-efficient fine-tuning, and cross-dataset generalization. Code is release at https://github.com/ChengShiest/Part2Object