NeurIPS 2019
Sun Dec 8th through Sat the 14th, 2019 at Vancouver Convention Center
Paper ID:3645
Title:Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds

This paper presents an instance segmentation algorithm on 3D point cloud data, which predicts 3D bounding box per instance and membership per point. All reviewers agree to the novelty and significance of the proposed method. This is a good piece of work and deserves to be accepted as a spotlight.