FoldingNet: Interpretable Unsupervised Learning on 3D Point Clouds
By Yaoqing Yang, Chen Feng, Yiru Shen, Dong Tian. 2017
84 citations recorded by the source corpus.
Recent deep networks that directly handle points in a point set, e.g., PointNet, have been state-of-the-art for supervised semantic 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 approach is proposed in the decoder, which folds a 2D grid onto the underlying 3D object surface of a point cloud. The proposed decoder only uses 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. Finally, this folding-based decoder is interpretable since the reconstruction could be viewed as a fine granular warping from the 2D grid to the point cloud surface.
Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, Geology
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Computational Mechanics
- Chi-Wing Fu
- P. Heng
- Daniel Cohen-Or
- Lequan Yu
- Andreas Geiger
- Martial Hebert
- David Held
- Christoph Mertz
In Computer Graphics and Computer-Aided Design
- P. Heng
- Chi-Wing Fu
- Daniel Cohen-Or
- Lequan Yu
- Andreas Geiger
- Martial Hebert
- David Held
- Christoph Mertz
In Geology
- Daniel Cohen-Or
- P. Heng
- Chi-Wing Fu
- Peter R. Florence
- Tanner Schmidt
- Martial Hebert
- Silvio Savarese
- Christoph Mertz
Citations in the corpus
- DeepSDF: Learning Continuous Signed Distance Functions for Shape Representation (2019, methodology): “To address this, recent data-driven approaches [57, 22] learn the parameterization task with deep networks.”
- EC-Net: an Edge-aware Point set Consolidation Network (2018, background): “The network was trained with an edge-aware joint loss function that has four terms; see the yellow boxes on the right. for high-level analysis problems with point clouds [31,32,33,34,35,36,37,38,39,40].”
- PCN: Point Completion Network (2018, methodology): “Recently, [60] introduces an interesting decoder design which mimics the deformation of a 2D plane into a 3D surface.”
- TopNet: Structural Point Cloud Decoder (2019, background): “] Folding-B[37] PCN-B[38] Ours(A) Ours(B)”
- Deep Local Shapes: Learning Local SDF Priors for Detailed 3D Reconstruction (2020, methodology): “The methods use generative point cloud models for scene representation [3,57,58].”