Shape Generation using Spatially Partitioned Point Clouds

Matheus Gadelha, Subhransu Maji and Rui Wang

Abstract

We propose a method to generate 3D shapes using point clouds. Given a point-cloud representation of a 3D shape, our method builds a kd-tree to spatially partition the points. This orders them consistently across all shapes, resulting in reasonably good correspondences across all shapes. We then use PCA analysis to derive a linear shape basis across the spatially partitioned points, and optimize the point ordering by iteratively minimizing the PCA reconstruction error. Even with the spatial sorting, the point clouds are inherently noisy and the resulting distribution over the shape coefficients can be highly multi-modal. We propose to use the expressive power of neural networks to learn a distribution over the shape coefficients in a generative-adversarial framework. Compared to 3D shape generative models trained on voxel-representations, our point-based method is considerably more light-weight and scalable, with little loss of quality. It also out-performs simpler linear factor models such as Probabilistic PCA, both qualitatively and quantitatively, on a number of categories from the ShapeNet dataset.

Session

Posters

Files

Paper (PDF)

DOI

10.5244/C.31.54
https://dx.doi.org/10.5244/C.31.54

Citation

Bibtex

            @inproceedings{BMVC2017_54,
                title={Shape Generation using Spatially Partitioned Point Clouds},
                author={Matheus Gadelha, Subhransu Maji and Rui Wang},
                year={2017},
                month={September},
                pages={54.1-54.12},
                articleno={54},
                numpages={12},
                booktitle={Proceedings of the British Machine Vision Conference (BMVC)},
                publisher={BMVA Press},
                editor={Tae-Kyun Kim, Stefanos Zafeiriou, Gabriel Brostow and Krystian Mikolajczyk},
                doi={10.5244/C.31.54},
                isbn={1-901725-60-X},
                url={https://dx.doi.org/10.5244/C.31.54}
            }