Pose-Guided 3D Human Generation in Indoor Scene

Authors

  • Minseok Kim UNIST
  • Changwoo Kang UNIST
  • Jeongin Park UNIST
  • Kyungdon Joo UNIST

DOI:

https://doi.org/10.1609/aaai.v37i1.25195

Keywords:

CV: 3D Computer Vision, ML: Deep Generative Models & Autoencoders

Abstract

In this work, we address the problem of scene-aware 3D human avatar generation based on human-scene interactions. In particular, we pay attention to the fact that physical contact between a 3D human and a scene (i.e., physical human-scene interactions) requires a geometrical alignment to generate natural 3D human avatar. Motivated by this fact, we present a new 3D human generation framework that considers geometric alignment on potential contact areas between 3D human avatars and their surroundings. In addition, we introduce a compact yet effective human pose classifier that classifies the human pose and provides potential contact areas of the 3D human avatar. It allows us to adaptively use geometric alignment loss according to the classified human pose. Compared to state-of-the-art method, our method can generate physically and semantically plausible 3D humans that interact naturally with 3D scenes without additional post-processing. In our evaluations, we achieve the improvements with more plausible interactions and more variety of poses than prior research in qualitative and quantitative analysis. Project page: https://bupyeonghealer.github.io/phin/.

Downloads

Published

2023-06-26

How to Cite

Kim, M., Kang, C., Park, J., & Joo, K. (2023). Pose-Guided 3D Human Generation in Indoor Scene. Proceedings of the AAAI Conference on Artificial Intelligence, 37(1), 1133-1141. https://doi.org/10.1609/aaai.v37i1.25195

Issue

Section

AAAI Technical Track on Computer Vision I