| CARVIEW |
Select Language
HTTP/2 301
server: GitHub.com
content-type: text/html
location: https://genforce.github.io/volumegan/
x-github-request-id: E330:272D88:80E4D8:908BBE:69519F8C
accept-ranges: bytes
age: 0
date: Sun, 28 Dec 2025 21:22:20 GMT
via: 1.1 varnish
x-served-by: cache-bom-vanm7210034-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766956941.625816,VS0,VE196
vary: Accept-Encoding
x-fastly-request-id: e9cbca52e752d97cc16529fd44d5a33c79c06527
content-length: 162
HTTP/2 200
server: GitHub.com
content-type: text/html; charset=utf-8
last-modified: Thu, 14 Jul 2022 04:19:11 GMT
access-control-allow-origin: *
strict-transport-security: max-age=31556952
etag: W/"62cf993f-1be8"
expires: Sun, 28 Dec 2025 21:32:20 GMT
cache-control: max-age=600
content-encoding: gzip
x-proxy-cache: MISS
x-github-request-id: C839:2B0FD4:80905C:903718:69519F8C
accept-ranges: bytes
date: Sun, 28 Dec 2025 21:22:21 GMT
via: 1.1 varnish
age: 0
x-served-by: cache-bom-vanm7210034-BOM
x-cache: MISS
x-cache-hits: 0
x-timer: S1766956941.835758,VS0,VE223
vary: Accept-Encoding
x-fastly-request-id: 44ef96571596463fae30f5fe2b378442bb3feea4
content-length: 2484
VolumeGAN
3D-aware Image Synthesis via Learning Structural and Textural Representations
1 The Chinese University of Hong Kong
2 Zhejiang University
3 ByteDance Inc.
Overview
This paper aims at achieving high-fidelity 3D-aware images synthesis.
We propose a novel framework, termed as VolumeGAN, for synthesizing images under different camera views, through explicitly learning a structural representation and a textural representation.
We first learn a feature volume to represent the underlying structure, which is then converted to a feature field using a NeRF-like model.
The feature field is further accumulated into a 2D feature map as the textural representation, followed by a neural renderer for appearance synthesis.
Such a design enables independent control of the shape and the appearance.
Extensive experiments on a wide range of datasets show that our approach achieves sufficiently higher image quality and better 3D control than the previous methods.
Results
Independent control of structure (shape) and texture (appearance) achieved by VolumeGAN.
Qualitative comparison between our VolumeGAN and existing alternatives.
 |
 |
Demo
BibTeX
@article{xu2021volumegan,
title = {3D-aware Image Synthesis via Learning Structural and Textural Representations},
author = {Xu, Yinghao and Peng, Sida and Yang, Ceyuan and Shen, Yujun and Zhou, Bolei},
article = {arXiv preprint arXiv:2112.10759},
year = {2021}
}
Related Work
Katja Schwarz, Yiyi Liao, Michael Niemeyer, Andreas Geiger.
GRAF: Generative Radiance Fields for 3D-Aware Image Synthesis.
NeurIPS, 2020.
Comment: Proposes the generative radiance fields for 3D-aware image synthesis.
Comment: Proposes the generative radiance fields for 3D-aware image synthesis.
Michael Niemeyer, Andreas Geiger.
GIRAFFE: Representing Scenes as Compositional Generative Neural Feature Fields.
CVPR, 2021.
Comment: Proposes the compositional generative neural feature fields for scene synthesis.
Comment: Proposes the compositional generative neural feature fields for scene synthesis.
Eric R. Chan, Marco Monteiro, Petr Kellnhofer, Jiajun Wu, Gordon Wetzstein.
pi-GAN: Periodic Implicit Generative Adversarial Networks for 3D-Aware Image Synthesis.
CVPR, 2021.
Comment: Proposes the periodic implicit generative neural feature fields for 3d-aware image synthesis.
Comment: Proposes the periodic implicit generative neural feature fields for 3d-aware image synthesis.
Comment: Proposes voxelized and implicit 3D representations and then render it to 2D image space with a reshape operation.