This study addresses the challenge of self-supervised learning for 3D mesh analysis. It presents an new approach that uses random walks as a form of data augmentation to generate diverse representations of mesh surfaces. Furthermore, it employs a combination of contrastive and clustering losses. The contrastive learning framework maximizes similarity between augmented instances of the same mesh while minimizing similarity between different meshes. We integrate this with a clustering loss, enhancing class distinction across training epochs and mitigating training variance. Our model's effectiveness is evaluated using mean Average Precision (mAP) scores and a supervised SVM linear classifier on extracted features, demonstrating its potential for various downstream tasks such as object classification and shape retrieval.
View on arXiv@article{yefet2025_2503.00816,
title={ Random Walks in Self-supervised Learning for Triangular Meshes },
author={ Gal Yefet and Ayellet Tal },
journal={arXiv preprint arXiv:2503.00816},
year={ 2025 }
}