Unsupervised Contrastive Representation Learning for 3D Mesh Segmentation (Student Abstract)
Ayaan Haque, Hankyu Moon, Heng Hao, Sima Didari, Jae Oh Woo, Patrick Bangert
Abstract
3D deep learning is a growing field of interest due to the vast amount of information stored in 3D formats. Triangular meshes are an efficient representation for irregular, non-uniform 3D objects. However, meshes are often challenging to annotate due to their high computational complexity. Therefore, it is desirable to train segmentation networks with limited-labeled data. Self-supervised learning (SSL), a form of unsupervised representation learning, is a growing alternative to fully-supervised learning which can decrease the burden of supervision for training. Specifically, contrastive learning (CL), a form of SSL, has recently been explored to solve limited-labeled data tasks. We propose SSL-MeshCNN, a CL method for pre-training CNNs for mesh segmentation. We take inspiration from prior CL frameworks to design a novel CL algorithm specialized for meshes. Our preliminary experiments show promising results in reducing the heavy labeled data requirement needed for mesh segmentation by at least 33%.
BibTeX
@article{Haque_Moon_Hao_Didari_Woo_Bangert_2024, title={Unsupervised Contrastive Representation Learning for 3D Mesh Segmentation (Student Abstract)}, volume={37}, url={https://ojs.aaai.org/index.php/AAAI/article/view/26971}, DOI={10.1609/aaai.v37i13.26971}, abstractNote={3D deep learning is a growing field of interest due to the vast amount of information stored in 3D formats. Triangular meshes are an efficient representation for irregular, non-uniform 3D objects. However, meshes are often challenging to annotate due to their high computational complexity. Therefore, it is desirable to train segmentation networks with limited-labeled data. Self-supervised learning (SSL), a form of unsupervised representation learning, is a growing alternative to fully-supervised learning which can decrease the burden of supervision for training. Specifically, contrastive learning (CL), a form of SSL, has recently been explored to solve limited-labeled data tasks. We propose SSL-MeshCNN, a CL method for pre-training CNNs for mesh segmentation. We take inspiration from prior CL frameworks to design a novel CL algorithm specialized for meshes. Our preliminary experiments show promising results in reducing the heavy labeled data requirement needed for mesh segmentation by at least 33%.}, number={13}, journal={Proceedings of the AAAI Conference on Artificial Intelligence}, author={Haque, Ayaan and Moon, Hankyu and Hao, Heng and Didari, Sima and Woo, Jae Oh and Bangert, Patrick}, year={2024}, month={Jul.}, pages={16222-16223} }