Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

EPContrast: Effective Point-level Contrastive Learning for Large-scale Point Cloud Understanding

Zhiyi Pan, Guoqing Liu, Wei Gao, Thomas H. Li

TL;DR

An Effective Pointlevel Contrastive Learning method for large-scale point cloud understanding dubbed EPContrast, which consists of AGContrast and ChannelContrast, which offers point-level contrastive loss while concurrently mitigating the computational resource burden.

Abstract

The acquisition of inductive bias through point-level contrastive learning holds paramount significance in point cloud pre-training. However, the square growth in computational requirements with the scale of the point cloud poses a substantial impediment to the practical deployment and execution. To address this challenge, this paper proposes an Effective Point-level Contrastive Learning method for large-scale point cloud understanding dubbed \textbf{EPContrast}, which consists of AGContrast and ChannelContrast. In practice, AGContrast constructs positive and negative pairs based on asymmetric granularity embedding, while ChannelContrast imposes contrastive supervision between channel feature maps. EPContrast offers point-level contrastive loss while concurrently mitigating the computational resource burden. The efficacy of EPContrast is substantiated through comprehensive validation on S3DIS and ScanNetV2, encompassing tasks such as semantic segmentation, instance segmentation, and object detection. In addition, rich ablation experiments demonstrate remarkable bias induction capabilities under label-efficient and one-epoch training settings.

EPContrast: Effective Point-level Contrastive Learning for Large-scale Point Cloud Understanding

TL;DR

An Effective Pointlevel Contrastive Learning method for large-scale point cloud understanding dubbed EPContrast, which consists of AGContrast and ChannelContrast, which offers point-level contrastive loss while concurrently mitigating the computational resource burden.

Abstract

The acquisition of inductive bias through point-level contrastive learning holds paramount significance in point cloud pre-training. However, the square growth in computational requirements with the scale of the point cloud poses a substantial impediment to the practical deployment and execution. To address this challenge, this paper proposes an Effective Point-level Contrastive Learning method for large-scale point cloud understanding dubbed \textbf{EPContrast}, which consists of AGContrast and ChannelContrast. In practice, AGContrast constructs positive and negative pairs based on asymmetric granularity embedding, while ChannelContrast imposes contrastive supervision between channel feature maps. EPContrast offers point-level contrastive loss while concurrently mitigating the computational resource burden. The efficacy of EPContrast is substantiated through comprehensive validation on S3DIS and ScanNetV2, encompassing tasks such as semantic segmentation, instance segmentation, and object detection. In addition, rich ablation experiments demonstrate remarkable bias induction capabilities under label-efficient and one-epoch training settings.

Paper Structure

This paper contains 14 sections, 5 equations, 2 figures, 4 tables.

Table of Contents

  1. Introduction
  2. Related Work
  3. Method
  4. Overview
  5. AGContrast
  6. ChannelContrast
  7. Loss Function
  8. Experiments
  9. Implementation
  10. Point Cloud Understanding
  11. Ablation Study
  12. One Epoch Training
  13. Label Efficient Learning
  14. Conclusion

Figures (2)

  • Figure 1: (Top) The semantic segmentation performance on Scratch, PointContrast, and our proposed EPContrast, and memory consumption on PointInfoNCE (used in PointContrast) and EPContrast. (Bottom) Visualization of point-level contrastive learning with increasing point cloud scale.
  • Figure 2: The pre-training framework with EPContrast, which consists of (a) AGContrast and (b) ChannelContrast.