App. Res. of Comp. Accepted Paper
Vol. 43, 2026 No. 7

Road driving scene generation method based on fusion of StyleGAN2 and self-attention mechanism

Zhao Chenxin1
Shao Jianwen2,3
Zhao Cunbin2,3
Wang Kai2,3
Xu Wenkan3
1. Hangzhou Polytechnic University, Hangzhou 310018, China
2. Zhejiang Key Laboratory of Digital Precision Measurement Technology Research, Hangzhou 310018, China
3. Zhejiang Institute of Quality Sciences, Hangzhou 310018, China
DOI: 10.19734/j.issn.1001-3695.2025.08.0447

Abstract

To address the limited quality of generated images caused by element diversity and structural complexity in complex road scenes, this paper proposed a road driving scene generation method fusing StyleGAN2 and self-attention mechanism. Firstly, based on road alignment and topological features, it divided driving scenes into four typical categories (one-way lanes, two-way lanes, intersections, ordinary curves) to reduce scene generation complexity. Secondly, it embedded multi-head self-attention modules into the generator and discriminator of StyleGAN2, and constructed a generator module architecture with cascaded style and attention modules. The self-attention mechanism was used to model global context of feature maps, enhancing the model’s ability to capture long-range dependencies—including road alignment continuity and complex spatial relationships between traffic elements. Experiments were conducted on the CIAC 2022 road dataset. Results show that the proposed method’s Fréchet Inception Distance (FID) on the unclassified dataset is significantly better than the original StyleGAN2 (with a 13.5% reduction) . On the classified dataset, performance improves further, with the one-way lane scene achieving the optimal FID (25.035) . Meanwhile, the method achieves a good balance between training efficiency and generation quality. This method effectively improves the structural rationality and visual realism of generated scenes, and provides an extensible data generation solution for autonomous driving simulation testing.

Foundation Support

国家重点研发计划课题(2022YFF0604803)
浙江省"尖兵"领雁"研发攻关计划项目(2023C01238)
浙江省基础公益研究计划项目(LGC22E050004)
浙江市市场监督管理局科技项目(QN2023426,CY2023106,ZD2024010)
杭州职业技术大学校级科研项目(KY202531,KY202527)

Publish Information

DOI: 10.19734/j.issn.1001-3695.2025.08.0447
Publish at: Application Research of Computers Accepted Paper, Vol. 43, 2026 No. 7

Publish History

[2026-03-25] Accepted Paper

Cite This Article

赵晨馨, 邵建文, 赵存彬, 等. 基于StyleGAN2和自注意力机制融合的道路驾驶场景生成方法 [J]. 计算机应用研究, 2026, 43 (7). (2026-03-27). https://doi.org/10.19734/j.issn.1001-3695.2025.08.0447. (Zhao Chenxin, Shao Jianwen, Zhao Cunbin, et al. Road driving scene generation method based on fusion of StyleGAN2 and self-attention mechanism [J]. Application Research of Computers, 2026, 43 (7). (2026-03-27). https://doi.org/10.19734/j.issn.1001-3695.2025.08.0447. )

About the Journal

  • Application Research of Computers Monthly Journal
  • Journal ID ISSN 1001-3695
    CN  51-1196/TP

Application Research of Computers, founded in 1984, is an academic journal of computing technology sponsored by Sichuan Institute of Computer Sciences under the Science and Technology Department of Sichuan Province.

Aiming at the urgently needed cutting-edge technology in this discipline, Application Research of Computers reflects the mainstream technology, hot technology and the latest development trend of computer application research at home and abroad in a timely manner. The main contents of the journal include high-level academic papers in this discipline, the latest scientific research results and major application results. The contents of the columns involve new theories of computer discipline, basic computer theory, algorithm theory research, algorithm design and analysis, blockchain technology, system software and software engineering technology, pattern recognition and artificial intelligence, architecture, advanced computing, parallel processing, database technology, computer network and communication technology, information security technology, computer image graphics and its latest hot application technology.

Application Research of Computers has many high-level readers and authors, and its readers are mainly senior and middle-level researchers and engineers engaged in the field of computer science, as well as teachers and students majoring in computer science and related majors in colleges and universities. Over the years, the total citation frequency and Web download rate of Application Research of Computers have been ranked among the top of similar academic journals in this discipline, and the academic papers published are highly popular among the readers for their novelty, academics, foresight, orientation and practicality.

Indexed & Evaluation

  • The Second National Periodical Award 100 Key Journals
  • Double Effect Journal of China Journal Formation
  • the Core Journal of China (Peking University 2023 Edition)
  • the Core Journal for Science
  • Chinese Science Citation Database (CSCD) Source Journals
  • RCCSE Chinese Core Academic Journals
  • Journal of China Computer Federation
  • 2020-2022 The World Journal Clout Index (WJCI) Report of Scientific and Technological Periodicals
  • Full-text Source Journal of China Science and Technology Periodicals Database
  • Source Journal of China Academic Journals Comprehensive Evaluation Database
  • Source Journals of China Academic Journals (CD-ROM Version), China Journal Network
  • 2017-2019 China Outstanding Academic Journals with International Influence (Natural Science and Engineering Technology)
  • Source Journal of Top Academic Papers (F5000) Program of China's Excellent Science and Technology Journals
  • Source Journal of China Engineering Technology Electronic Information Network and Electronic Technology Literature Database
  • Source Journal of British Science Digest (INSPEC)
  • Japan Science and Technology Agency (JST) Source Journal
  • Russian Journal of Abstracts (AJ, VINITI) Source Journals
  • Full-text Journal of EBSCO, USA
  • Cambridge Scientific Abstracts (Natural Sciences) (CSA(NS)) core journals
  • Poland Copernicus Index (IC)
  • Ulrichsweb (USA)