Point Cloud Instance Segmentation using Probabilistic Embeddings

Abstract

In this paper we propose a new framework for point cloud instance segmentation. Our framework has two steps: an embedding step and a clustering step. In the embedding step, our main contribution is to propose a probabilistic embedding space for point cloud embedding. Specifically, each point is represented as a tri-variate normal distribution. In the clustering step, we propose a novel loss function, which benefits both the semantic segmentation and the clustering. Our experimental results show important improvements to the SOTA, i.e., 3.1% increased average per-category mAP on the PartNet dataset.

arXiv commentAccepted by CVPR 2021. Project: http://1zb.github.io/publication/prob-embed/

References in corpus (11)

• Adam: A Method for Stochastic Optimization
• Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
• PyTorch: An Imperative Style, High-Performance Deep Learning Library
• What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?
• Stochastic Backpropagation and Approximate Inference in Deep Generative Models
• Fast Graph Representation Learning with PyTorch Geometric
• Semantic Instance Segmentation with a Discriminative Loss Function
• Semantic Instance Segmentation via Deep Metric Learning
• 3D Segmentation with Exponential Logarithmic Loss for Highly Unbalanced Object Sizes
• Striking the Right Balance with Uncertainty
• Dynamic Scale Inference by Entropy Minimization

Cited by in corpus (5)

• SalsaNext: Fast, Uncertainty-aware Semantic Segmentation of LiDAR Point Clouds for Autonomous Driving
• Deep Learning for 3D Point Clouds: A Survey
• Deep Learning for 3D Point Cloud Understanding: A Survey
• Instance Segmentation in 3D Scenes using Semantic Superpoint Tree Networks
• Scan2Part: Fine-grained and Hierarchical Part-level Understanding of Real-World 3D Scans