PointLIE: Locally Invertible Embedding for Point Cloud Sampling and Recovery

Zhao, Weibing; Yan, Xu; Gao, Jiantao; Zhang, Ruimao; Zhang, Jiayan; Li, Zhen; Wu, Song; Cui, Shuguang

doi:10.24963/ijcai.2021/186

PointLIE: Locally Invertible Embedding for Point Cloud Sampling and Recovery

Weibing Zhao, Xu Yan, Jiantao Gao, Ruimao Zhang, Jiayan Zhang, Zhen Li, Song Wu, Shuguang Cui

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 1345-1351.

https://doi.org/10.24963/ijcai.2021/186

Keywords

Computer Vision: 2D and 3D Computer Vision
Machine Learning: Deep Learning

Abstract

Point Cloud Sampling and Recovery (PCSR) is critical for massive real-time point cloud collection and processing since raw data usually requires large storage and computation. This paper addresses a fundamental problem in PCSR: How to downsample the dense point cloud with arbitrary scales while preserving the local topology of discarded points in a case-agnostic manner (i.e., without additional storage for point relationships)? We propose a novel Locally Invertible Embedding (PointLIE) framework to unify the point cloud sampling and upsampling into one single framework through bi-directional learning. Specifically, PointLIE decouples the local geometric relationships between discarded points from the sampled points by progressively encoding the neighboring offsets to a latent variable. Once the latent variable is forced to obey a pre-defined distribution in the forward sampling path, the recovery can be achieved effectively through inverse operations. Taking the recover-pleasing sampled points and a latent embedding randomly drawn from the specified distribution as inputs, PointLIE can theoretically guarantee the fidelity of reconstruction and outperform state-of-the-arts quantitatively and qualitatively.

Weibing Zhao, Xu Yan, Jiantao Gao, Ruimao Zhang, Jiayan Zhang, Zhen Li, Song Wu, and Shuguang Cui. 2021. PointLIE: Locally Invertible Embedding for Point Cloud Sampling and Recovery. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 1345–1351. https://doi.org/10.24963/ijcai.2021/186

@inproceedings{ijcai2021p186,
  title     = {PointLIE: Locally Invertible Embedding for Point Cloud Sampling and Recovery},
  author    = {Zhao, Weibing and Yan, Xu and Gao, Jiantao and Zhang, Ruimao and Zhang, Jiayan and Li, Zhen and Wu, Song and Cui, Shuguang},
  booktitle = {Proceedings of the Thirtieth International Joint Conference on
               Artificial Intelligence, {IJCAI-21}},
  publisher = {International Joint Conferences on Artificial Intelligence Organization},
  editor    = {Prof. Zhi-Hua Zhou},
  pages     = {1345--1351},
  year      = {2021},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2021/186},
  url       = {https://doi.org/10.24963/ijcai.2021/186},
}

TY  - CPAPER
TI  - PointLIE: Locally Invertible Embedding for Point Cloud Sampling and Recovery
AU  - Zhao, Weibing
AU  - Yan, Xu
AU  - Gao, Jiantao
AU  - Zhang, Ruimao
AU  - Zhang, Jiayan
AU  - Li, Zhen
AU  - Wu, Song
AU  - Cui, Shuguang
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 1345
EP  - 1351
DO  - 10.24963/ijcai.2021/186
UR  - https://doi.org/10.24963/ijcai.2021/186
ER  -