arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2203.00680cs.CV
收藏

CrossPoint:自监督跨模态对比学习三维点云理解

CrossPoint: Self-Supervised Cross-Modal Contrastive Learning for 3D Point Cloud Understanding

The University of Sydney(悉尼大学)

Mohamed Afham, Isuru Dissanayake, Dinithi Dissanayake, Amaya Dharmasiri, Kanchana Thilakarathna, Ranga Rodrigo

摘要:

由于点云数据结构的不规则性,对大规模点云数据集进行人工标注,如3D目标分类、分割和检测,往往是一项非常费力的工作。自监督学习不需要任何人工标记,是解决这个问题的一种很有前途的方法。我们在现实世界中观察到,人类能够映射从2D图像中学习到的视觉概念来理解3D世界。受此启发,我们提出了CrossPoint,一种简单的跨模态对比学习方法来学习可转移的3D点云表示。它通过最大化点云与不变空间中对应的渲染2D图像之间的一致性来实现对象的3D-2D对应,同时鼓励点云模态中的变换的不变性。我们的联合训练目标结合了模态内和模态间的特征对应关系,从而以自监督的方式集成了来自3D点云和2D图像模态的丰富学习信号。实验结果表明,我们的方法优于以前的无监督学习方法在各种下游任务,包括3D对象分类和分割。此外,消融研究验证了我们的方法的潜力,更好地了解点云。代码和预训练模型可在http://github.com/MohamedAfham/CrossPoint上获得。

英文摘要:

Manual annotation of large-scale point cloud dataset for varying tasks such as 3D object classification, segmentation and detection is often laborious owing to the irregular structure of point clouds. Self-supervised learning, which operates without any human labeling, is a promising approach to address this issue. We observe in the real world that humans are capable of mapping the visual concepts learnt from 2D images to understand the 3D world. Encouraged by this insight, we propose CrossPoint, a simple cross-modal contrastive learning approach to learn transferable 3D point cloud representations. It enables a 3D-2D correspondence of objects by maximizing agreement between point clouds and the corresponding rendered 2D image in the invariant space, while encouraging invariance to transformations in the point cloud modality. Our joint training objective combines the feature correspondences within and across modalities, thus ensembles a rich learning signal from both 3D point cloud and 2D image modalities in a self-supervised fashion. Experimental results show that our approach outperforms the previous unsupervised learning methods on a diverse range of downstream tasks including 3D object classification and segmentation. Further, the ablation studies validate the potency of our approach for a better point cloud understanding. Code and pretrained models are available at http://github.com/MohamedAfham/CrossPoint.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑