Multi-to-Single Knowledge Distillation for Point Cloud Semantic Segmentation

3D point cloud semantic segmentation is one of the fundamental tasks for environmental understanding. Although significant progress has been made in recent years, the performance of classes with few examples or few points is still far from satisfactory. In this paper, we propose a novel multi-to-single knowledge distillation framework for the 3D point cloud semantic segmentation task to boost the performance of those hard classes. Instead of fusing all the points of multi-scans directly, only the instances that belong to the previously defined hard classes are fused. To effectively and sufficiently distill valuable knowledge from multi-scans, we leverage a multilevel distillation framework, i. e. , feature representation distillation, logit distillation, and affinity distillation. We further develop a novel instance-aware affinity distillation algorithm for capturing high-level structural knowledge to enhance the distillation efficacy for hard classes. Finally, we conduct experiments on the SemanticKITTI dataset, and the results on both the validation and test sets demonstrate that our method yields substantial improvements compared with the baseline method. The code is available at https://github.com/skyshoumeng/M2SKD.

Authors

• Shoumeng Qiu
• Feng Jiang
• Haiqiang Zhang
• Xiangyang Xue 0001
• Jian Pu

Keywords

• Point cloud compression
• Training
• Three-dimensional displays
• Codes
• Automation
• Fuses
• Semantic segmentation
• Point Cloud
• Point Cloud Semantic Segmentation
• Classification Performance
• Feature Representation
• Baseline Methods
• Segmentation Task
• 3D Segmentation
• Multilevel Framework
• Semantic Segmentation Task
• 3D Tasks
• Sequence Information
• Sparsity
• Object Detection
• Teacher Model
• Kullback-Leibler
• Similarity Matrix
• Fusion Method
• Segmentation Performance
• Frequent Class
• Student Model
• Point Cloud Registration
• Fusion Operation
• Distillation Process
• Fusion Results
• Semantic Labels
• 3D Patterns
• Global Translation
• Sparse Point Cloud
• Distillation Loss