Unsupervised 3D point cloud segmentation using multicut graph algorithm on features with FreePoint
Unsupervised 3D point cloud segmentation using multicut graph algorithm on features with FreePoint
FreePoint: Unsupervised Point Cloud Instance Segmentation
arXiv paper abstract https://arxiv.org/abs/2305.06973
arXiv PDF paper https://arxiv.org/pdf/2305.06973.pdf
Instance segmentation of point clouds is a crucial task in 3D field ... However ... requires a large number of manual annotations
... propose a method, called FreePoint, for underexplored unsupervised class-agnostic instance segmentation on point clouds.
... represent the point features by combining coordinates, colors, normals, and self-supervised deep features.
Based on the point features, ... perform a multicut algorithm to segment point clouds into coarse instance masks as pseudo labels, which are used to train a point cloud instance segmentation model.
... propose a weakly-supervised training strategy and corresponding loss.
... For class-agnostic instance segmentation on point clouds, FreePoint largely fills the gap with its fully-supervised counterpart based on the state-of-the-art instance segmentation model Mask3D and even surpasses some previous fully-supervised methods ...
Please like and share this post if you enjoyed it using the buttons at the bottom!
Stay up to date. Subscribe to my posts https://morrislee1234.wixsite.com/website/contact
Web site with my other posts by category https://morrislee1234.wixsite.com/website
Comments