TR2018-042

FoldingNet: Point Cloud Auto-encoder via Deep Grid Deformation


Abstract:

Recent deep networks that directly handle points in a point set, e.g., PointNet, have been state-of-the-art for supervised learning tasks on point clouds such as classification and segmentation. In this work, a novel end-to-end deep auto-encoder is proposed to address unsupervised learning challenges on point clouds. On the encoder side, a graph-based enhancement is enforced to promote local structures on top of PointNet. Then, a novel folding-based decoder deforms a canonical 2D grid onto the underlying 3D object surface of a point cloud, achieving low reconstruction errors even for objects with delicate structures. The proposed decoder only use about 7% parameters of a decoder with fully-connected neural networks, yet leads to a more discriminative representation that achieves higher linear SVM classification accuracy than the benchmark. In addition, the proposed decoder structure is shown, in theory, to be a generic architecture that is able to reconstruct an arbitrary point cloud from a 2D grid. Our code is available at http://www.merl.com/research/license#FoldingNet

 

  • Software & Data Downloads

  • Related Publication

  •  Yang, Y., Feng, C., Shen, Y., Tian, D., "FoldingNet: Interpretable Unsupervised Learning on 3D Point Clouds", arXiv, December 2017.
    BibTeX arXiv Software
    • @article{Yang2017dec,
    • author = {Yang, Yaoqing and Feng, Chen and Shen, Yiru and Tian, Dong},
    • title = {FoldingNet: Interpretable Unsupervised Learning on 3D Point Clouds},
    • journal = {arXiv},
    • year = 2017,
    • month = dec,
    • url = {https://arxiv.org/abs/1712.07262}
    • }