Unsupervised Contrastive Representation Learning for 3D Mesh Segmentation (Student Abstract)


  • Ayaan Haque University of California, Berkeley Samsung SDS Research America
  • Hankyu Moon Samsung SDS Research America
  • Heng Hao Samsung SDS Research America
  • Sima Didari Samsung SDS Research America
  • Jae Oh Woo Samsung SDS Research America
  • Patrick Bangert Samsung SDS Research America




Contrastive Learning, Unsupervised Learning, Self-Supervised Learning, 3D Mesh, Segmentation


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%.




How to Cite

Haque, A., Moon, H., Hao, H., Didari, S., Woo, J. O., & Bangert, P. (2023). Unsupervised Contrastive Representation Learning for 3D Mesh Segmentation (Student Abstract). Proceedings of the AAAI Conference on Artificial Intelligence, 37(13), 16222-16223. https://doi.org/10.1609/aaai.v37i13.26971