arrow
返回

DeformSg2im: Scene graph based multi-instance image generation with a deformable geometric layout

delete2023-11-01
delete1
PRE
AI
Z
Zhiming Wang
Y
Yuxiao Li
D
Danlan Huang
王
王隽 (Juan Wang)
N
Ning Ge *
J
Jianhua Lu
DOI:10.1016/j.neucom.2023.126684delete
delete原文链接
delete原文求助
delete分享
delete收藏
摘要

摘要

En 中文
Recently, scene graph based image generation has emerged to be an important research direction for advanced multi-instance generation tasks. Scene layout generation, the phase that produces the instance-wise visual representations and maintains the spatial relationships among all instances, is vital in translating a scene graph to an image. However, the scene layouts generated by existing methods are too coarse and semantically inconsistent with the given scene graph, which results in quality degradation of the generated images. Motivated by this, we propose a novel scene graph based image generation model called DeformSg2im, which aims to generate visually appealing and semantically faithful images according to the given scene graphs. Our method addresses the aforementioned problems from two aspects. In one aspect, we present an attention-based instance embedding estimator to refine the shape information for each instance. By introducing attention maps to the estimation of instance embeddings, our method is able to generate more plausible instances with sharp edges on the images. In the other aspect, a spatial warping network (SWN) is proposed to adaptively capture the spatial dependencies among instances. With sequential modeling and geometric deformations, the SWN is capable of generating the scene layout conforming to the given scene graph. Extensive experiments show that our model generates images with high visual quality and achieves competitive quantitative results compared to existing works. Ablation studies on the proposed modules are also conducted, and the results demonstrate the effectiveness of our method.
Keyword:
Scene graph
Multi-instance image generation
Attention-based mask synthesis
Sequential modeling
Geometric deformation

期刊

Neurocomputing 封面图
Neurocomputing
IF:
6.5
论文数:
2.5W
被引数:
6.5W

机构

T
tsinghua university
学者数:
11.9W
论文数: 10.0W
被引数: 137
I
institute of automation, cas
学者数:
2.2K
论文数: 2.1K
被引数: 2
引用论文

引用论文

err分享
err收藏
Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations视觉基因组: 使用众包密集图像注释连接语言和视觉
err2017-02-06
err3.1K
errOAAI
errKrishna, Ranjay; Zhu, Yuke; Groth, Oliver; Johnson, Justin; Hata, Kenji; Kravitz, Joshua; Chen, Stephanie; Kalantidis, Yannis; Li, Li-Jia; Shamma, David A.; Bernstein, Michael S.; Li Fei-Fei
err分享
err收藏
Without shelter, people die: disproportionate mortality among King County’s homeless population, 2009–2019
err2022-01-18
err0
PREAI
errRachel Scott; Michele Marchand; Bert Stover; Kate Causey; Richard Harruff; Amy Hagopian
err分享
err收藏
Infrared focal plane array incorporating silicon IC process compatible bolometer
err1996-01-01
err0
PREAI
errA. Tanaka; S. Matsumoto; N. Tsukamoto; S. Itoh; K. Chiba; T. Endoh; A. Nakazato; K. Okuyama; Y. Kumazawa; M. Hijikawa; H. Gotoh; T. Tanaka; N. Teranishi
err分享
err收藏
err分享
err收藏
学者 查看更多内容