aiqbench analysis
The Dynamic SpectraFormer addresses critical visibility issues for Autonomous Underwater Vehicles by leveraging frequency domain transformers to simultaneously correct low-frequency color distortions and high-frequency texture details. This approach represents significant progress in computer vision applications specifically tailored for robotics operating in challenging aquatic environments where traditional spatial-domain methods fall short.
Dynamic SpectraFormer for Ultra-High-Definition Underwater Image Enhancement
arXiv:2608.18662v1 Announce Type: cross Abstract: Underwater images suffer from color distortion, haze, and poor visibility due to light refraction and absorption in water. These challenges significantly impact the utilization of Autonomous Underwater Vehicles (AUVs) or marine robots. Typically, color and brightness distortions manifest at lower frequencies, while edge and texture distortions are prevalent at higher frequencies. Traditional methods struggle to concurrently rectify these mixed distortions as they primarily concentrate on the spatial domain. To address these issues, we introduce the Dynamic SpectraFormer, which enhances underwater images through a frequency domain transformer. The Dynamic SpectraFormer introduces an ultra-high-resolution sparse spectrum attention module, which could capture the long-term dependency without losing the universal approximating power. Additionally, we have developed a dynamic spectrum weight generation layer that serves as an adaptive spectrum band selector, accentuating critical frequency bands and suppressing less relevant ones. Consequently, this method significantly improves underwater image quality by addressing both high- and low-frequency distortions. Our extensive ablation studies and comparative evaluations consolidate the Dynamic SpectraFormer's efficacy across multiple underwater image enhancement benchmarks. The source code is available at https://github.com/arifence2024/DynamicSpectraFormer.git.
Abstract:Underwater images suffer from color distortion, haze, and poor visibility due to light refraction and absorption in water. These challenges significantly impact the utilization of Autonomous Underwater Vehicles (AUVs) or marine robots. Typically, color and brightness distortions manifest at lower frequencies, while edge and texture distortions are prevalent at higher frequencies. Traditional methods struggle to concurrently rectify these mixed distortions as they primarily concentrate on the spatial domain. To address these issues, we introduce the Dynamic SpectraFormer, which enhances underwater images through a frequency domain transformer. The Dynamic SpectraFormer introduces an ultra-high-resolution sparse spectrum attention module, which could capture the long-term dependency without losing the universal approximating power. Additionally, we have developed a dynamic spectrum weight generation layer that serves as an adaptive spectrum band selector, accentuating critical frequency bands and suppressing less relevant ones. Consequently, this method significantly improves underwater image quality by addressing both high- and low-frequency distortions. Our extensive ablation studies and comparative evaluations consolidate the Dynamic SpectraFormer's efficacy across multiple underwater image enhancement benchmarks. The source code is available at this https URL.
8 pages, 7 figures. Published in the 2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2024)
Computer Vision and Pattern Recognition (cs.CV); Robotics (cs.RO)
arXiv:2608.18662 [cs.CV]
(or arXiv:2608.18662v1 [cs.CV] for this version)
https://doi.org/10.48550/arXiv.2608.18662 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
2024 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), pp. 8634-8641, 2024
https://doi.org/10.1109/IROS58592.2024.10802529 Focus to learn more DOI(s) linking to related resources
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 317 words
