Learning Gradient Fields for Shape Generation

Publikation: Bidrag til tidsskriftKonferenceartikelForskningfagfællebedømt

  • Ruojin Cai
  • Guandao Yang
  • Hadar Averbuch-Elor
  • Zekun Hao
  • Belongie, Serge
  • Noah Snavely
  • Bharath Hariharan

In this work, we propose a novel technique to generate shapes from point cloud data. A point cloud can be viewed as samples from a distribution of 3D points whose density is concentrated near the surface of the shape. Point cloud generation thus amounts to moving randomly sampled points to high-density areas. We generate point clouds by performing stochastic gradient ascent on an unnormalized probability density, thereby moving sampled points toward the high-likelihood regions. Our model directly predicts the gradient of the log density field and can be trained with a simple objective adapted from score-based generative models. We show that our method can reach state-of-the-art performance for point cloud auto-encoding and generation, while also allowing for extraction of a high-quality implicit surface. Code is available at https://github.com/RuojinCai/ShapeGF.

OriginalsprogEngelsk
TidsskriftLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Sider (fra-til)364-381
Antal sider18
ISSN0302-9743
DOI
StatusUdgivet - 2020
Eksternt udgivetJa
Begivenhed16th European Conference on Computer Vision, ECCV 2020 - Glasgow, Storbritannien
Varighed: 23 aug. 202028 aug. 2020

Konference

Konference16th European Conference on Computer Vision, ECCV 2020
LandStorbritannien
ByGlasgow
Periode23/08/202028/08/2020

Bibliografisk note

Publisher Copyright:
© 2020, Springer Nature Switzerland AG.

ID: 301818367