Five A+ Network: You Only Need 9K Parameters for Underwater Image Enhancement
JingXia Jiang (jimei university), Tian Ye (The Hong Kong University of Science and Technology (Guangzhou)),* Sixiang Chen (The Hong Kong University of Science and Technology (Guangzhou)), Erkang Chen (Jimei University), Yun Liu (Southwest University), Shi Jun (XinJiang University), Jinbin Bai (Nanjing University), Wenhao Chai (University of Washington)
The 34th British Machine Vision Conference
Abstract
A lightweight underwater image enhancement network is of great significance for resource-constrained platforms, but balancing model size, computational efficiency, and enhancement performance has proven difficult for previous approaches. In this work, we propose the Five A+ Network (FA+ Net), a highly efficient and lightweight real-time underwater image enhancement network with only ~ 9k parameters and ~ 0.01s processing time. The FA+ Net employs a two-stage enhancement structure. The powerful prior stage aims to decompose challenging underwater degradations into sub-problems, while the fine-grained stage incorporates multi-branch color enhancement module and pixel attention module to amplify the network’s perception of details. To the best of our knowledge, FA+ Net is the only network with the capability of real-time enhancement of 1080P images. Through extensive experiments and comprehensive visual comparisons, we show that FA+ Net outperforms previous approaches by obtaining state-of-the-art performance on multiple datasets while significantly reducing both the number of parameters and computational complexity. The code is available at https://github.com/Owen718/FiveAPlus-Network.Video
Citation
@inproceedings{Jiang_2023_BMVC,
author = {JingXia Jiang and Tian Ye and Sixiang Chen and Erkang Chen and Yun Liu and Shi Jun and Jinbin Bai and Wenhao Chai},
title = {Five A+ Network: You Only Need 9K Parameters for Underwater Image Enhancement},
booktitle = {34th British Machine Vision Conference 2023, {BMVC} 2023, Aberdeen, UK, November 20-24, 2023},
publisher = {BMVA},
year = {2023},
url = {https://papers.bmvc2023.org/0149.pdf}
}
Copyright © 2023
The British Machine Vision Conference is organised by