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

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Pan, Zhiyi, Liu, Guoqing, Gao, Wei, Li, Thomas H.
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929554451333120
author Pan, Zhiyi
Liu, Guoqing
Gao, Wei
Li, Thomas H.
author_facet Pan, Zhiyi
Liu, Guoqing
Gao, Wei
Li, Thomas H.
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2410_17207
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle EPContrast: Effective Point-level Contrastive Learning for Large-scale Point Cloud Understanding
Pan, Zhiyi
Liu, Guoqing
Gao, Wei
Li, Thomas H.
Computer Vision and Pattern Recognition
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.
title EPContrast: Effective Point-level Contrastive Learning for Large-scale Point Cloud Understanding
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2410.17207