18‐2: Research on Lightweight Super‐Resolution GAN Model on Low‐Cost FPGA for Automotive Application
SID Symposium Digest of Technical Papers · 2026-08 · 57 (S1) · 165-168
DOI: 10.1002/sdtp.70003_41 ↗Kurz gefasst
Fachartikel in der Zeitschrift SID Symposium Digest of Technical Papers, online erschienen am 11.09.2026. Verfasst von Zhang Jiahe und 1 weiteren Autor:innen. Die Arbeit nennt 6 Literaturangaben.
Abstract (Originalsprache)
The contradiction between the demand for low‐resolution (LR) network video enhancement on automotive high‐resolution displays and the hardware constraints of low‐cost FPGAs. This paper proposes a structured pruning scheme for a generative adversarial network for image super resolution (SRGAN) adapted to Xilinx Artix‐7 7015, completing model training and deployment. The parameter count is reduced by 52%, with inference latency less 28ms/frame. PSNR is improved by 2.3dB compared with bicubic interpolation, meeting the real‐time requirements of automotive scenarios.
Zitieren
APA
Jiahe, Z., & Lili, W. (2026). 18‐2: Research on Lightweight Super‐Resolution GAN Model on Low‐Cost FPGA for Automotive Application. SID Symposium Digest of Technical Papers, 57(S1), 165-168. https://doi.org/10.1002/sdtp.70003_41
Vancouver
Jiahe Z, Lili W. 18‐2: Research on Lightweight Super‐Resolution GAN Model on Low‐Cost FPGA for Automotive Application. SID Symposium Digest of Technical Papers. 2026;57(S1):165-168. doi:10.1002/sdtp.70003_41
BibTeX
@article{Jiahe2026,
author = {Jiahe, Zhang and Lili, Wang},
title = {18‐2: Research on Lightweight Super‐Resolution GAN Model on Low‐Cost FPGA for Automotive Application},
journal = {SID Symposium Digest of Technical Papers},
year = {2026},
volume = {57},
number = {S1},
pages = {165--168},
doi = {10.1002/sdtp.70003_41}
}
EUVIMED