arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2308.03381cs.CV
收藏

面向低光视觉的双层生成学习

Bilevel Generative Learning for Low-Light Vision

Dalian University of Technology(大连理工大学);Peng Cheng Laboratory(鹏城实验室)

Yingchi Liu, Zhu Liu, Long Ma, Jinyuan Liu, Xin Fan, Zhongxuan Luo, Risheng Liu

AI总结:

本文提出一种基于双层优化的通用低光视觉方法,通过RAW到RGB的生成块连接增强、检测和分割任务,并设计低成本与高质量两种学习策略,实验验证了其优越性。

摘要:

近年来,人们对构建用于低光视觉(LLV)的深度学习方案越来越感兴趣。现有技术主要侧重于在标准RGB域上设计任务特定且数据依赖的视觉模型,这些模型本质上包含潜在的数据关联。在本研究中,我们提出了一种通用的低光视觉解决方案,通过引入一个生成块将数据从RAW域转换到RGB域。这种新颖的方法通过显式刻画数据生成过程,将不同的视觉问题连接起来,这在该领域尚属首次。为了精确刻画生成过程与视觉任务之间的潜在对应关系,我们建立了一个双层模型,其中生成块的参数定义为上层,视觉任务的参数定义为下层。我们进一步开发了两种针对不同目标(即低成本和高质量)的学习策略,以获得一种新的双层生成学习范式。通过在增强任务上进行双层优化,生成块在其他低光视觉任务中具有很强的泛化能力。在三个具有代表性的低光视觉任务(即增强、检测和分割)上的大量实验评估充分证明了我们所提出方法的优越性。代码将公开于 https://github.com/Yingchi1998/BGL。

英文摘要:

Recently, there has been a growing interest in constructing deep learning schemes for Low-Light Vision (LLV). Existing techniques primarily focus on designing task-specific and data-dependent vision models on the standard RGB domain, which inherently contain latent data associations. In this study, we propose a generic low-light vision solution by introducing a generative block to convert data from the RAW to the RGB domain. This novel approach connects diverse vision problems by explicitly depicting data generation, which is the first in the field. To precisely characterize the latent correspondence between the generative procedure and the vision task, we establish a bilevel model with the parameters of the generative block defined as the upper level and the parameters of the vision task defined as the lower level. We further develop two types of learning strategies targeting different goals, namely low cost and high accuracy, to acquire a new bilevel generative learning paradigm. The generative blocks embrace a strong generalization ability in other low-light vision tasks through the bilevel optimization on enhancement tasks. Extensive experimental evaluations on three representative low-light vision tasks, namely enhancement, detection, and segmentation, fully demonstrate the superiority of our proposed approach. The code will be available at https://github.com/Yingchi1998/BGL.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑