Unsupervised Point Cloud Pre-Training via Contrasting and Clustering
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2022
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908886404956160 |
|---|---|
| author | Mei, Guofeng Huang, Xiaoshui Liu, Juan Zhang, Jian Wu, Qiang |
| author_facet | Mei, Guofeng Huang, Xiaoshui Liu, Juan Zhang, Jian Wu, Qiang |
| contents | Annotating large-scale point clouds is highly time-consuming and often infeasible for many complex real-world tasks. Point cloud pre-training has therefore become a promising strategy for learning discriminative representations without labeled data. In this paper, we propose a general unsupervised pre-training framework, termed ConClu, which jointly integrates contrasting and clustering. The contrasting objective maximizes the similarity between feature representations extracted from two augmented views of the same point cloud, while the clustering objective simultaneously partitions the data and enforces consistency between cluster assignments across augmentations. Experimental results on multiple downstream tasks show that our method outperforms state-of-the-art approaches, demonstrating the effectiveness of the proposed framework. Code is available at https://github.com/gfmei/conclu. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2202_02543 |
| institution | arXiv |
| publishDate | 2022 |
| record_format | arxiv |
| spellingShingle | Unsupervised Point Cloud Pre-Training via Contrasting and Clustering Mei, Guofeng Huang, Xiaoshui Liu, Juan Zhang, Jian Wu, Qiang Computer Vision and Pattern Recognition Artificial Intelligence Optimization and Control Annotating large-scale point clouds is highly time-consuming and often infeasible for many complex real-world tasks. Point cloud pre-training has therefore become a promising strategy for learning discriminative representations without labeled data. In this paper, we propose a general unsupervised pre-training framework, termed ConClu, which jointly integrates contrasting and clustering. The contrasting objective maximizes the similarity between feature representations extracted from two augmented views of the same point cloud, while the clustering objective simultaneously partitions the data and enforces consistency between cluster assignments across augmentations. Experimental results on multiple downstream tasks show that our method outperforms state-of-the-art approaches, demonstrating the effectiveness of the proposed framework. Code is available at https://github.com/gfmei/conclu. |
| title | Unsupervised Point Cloud Pre-Training via Contrasting and Clustering |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Optimization and Control |
| url | https://arxiv.org/abs/2202.02543 |