Spline Positional Encoding for Learning 3D Implicit Signed Distance Fields

Wang, Peng-Shuai; Liu, Yang; Yang, Yu-Qi; Tong, Xin

doi:10.24963/ijcai.2021/151

Spline Positional Encoding for Learning 3D Implicit Signed Distance Fields

Peng-Shuai Wang, Yang Liu, Yu-Qi Yang, Xin Tong

Download BibTeX Cite

Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence. Main Track. Pages 1091-1097.

https://doi.org/10.24963/ijcai.2021/151

Keywords

Computer Vision: 2D and 3D Computer Vision

Abstract

Multilayer perceptrons (MLPs) have been successfully used to represent 3D

shapes implicitly and compactly, by mapping 3D coordinates to the

corresponding signed distance values or occupancy values. In this paper, we

propose a novel positional encoding scheme, called Spline Positional

Encoding, to map the input coordinates to a high dimensional space before

passing them to MLPs, which help recover 3D signed distance fields with

fine-scale geometric details from unorganized 3D point clouds. We verified

the superiority of our approach over other positional encoding schemes on

tasks of 3D shape reconstruction and 3D shape space learning from input

point clouds. The efficacy of our approach extended to image reconstruction

is also demonstrated and evaluated.

Peng-Shuai Wang, Yang Liu, Yu-Qi Yang, and Xin Tong. 2021. Spline Positional Encoding for Learning 3D Implicit Signed Distance Fields. In Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21. IJCAI, 1091–1097. https://doi.org/10.24963/ijcai.2021/151

@inproceedings{ijcai2021p151,
  title     = {Spline Positional Encoding for Learning 3D Implicit Signed Distance Fields},
  author    = {Wang, Peng-Shuai and Liu, Yang and Yang, Yu-Qi and Tong, Xin},
  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     = {1091--1097},
  year      = {2021},
  month     = {8},
  note      = {Main Track},
  doi       = {10.24963/ijcai.2021/151},
  url       = {https://doi.org/10.24963/ijcai.2021/151},
}

TY  - CPAPER
TI  - Spline Positional Encoding for Learning 3D Implicit Signed Distance Fields
AU  - Wang, Peng-Shuai
AU  - Liu, Yang
AU  - Yang, Yu-Qi
AU  - Tong, Xin
T2  - Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, IJCAI-21
PB  - International Joint Conferences on Artificial Intelligence Organization
PY  - 2021
SP  - 1091
EP  - 1097
DO  - 10.24963/ijcai.2021/151
UR  - https://doi.org/10.24963/ijcai.2021/151
ER  -