Abstract
By contrast with the prevailing works of low-light enhancement in natural images and videos, this study copes with the low-illumination quality degradation in anime scenery images to bridge the domain gap. For such an underexplored enhancement task, we first curate images from various sources and construct an unpaired anime scenery dataset with diverse environments and illumination conditions to address the data scarcity. To exploit the power of uncertainty information inherent with the diverse illumination conditions, we propose a Data Relativistic Uncertainty (DRU) framework, motivated by the idea from Relativistic GAN. By analogy with the wave-particle duality of light, our framework interpretably defines and quantifies the illumination uncertainty of dark/bright samples, which is leveraged to dynamically adjust the objective functions to recalibrate the model learning under data uncertainty. Extensive experiments demonstrate the effectiveness of DRU framework by training several versions of EnlightenGANs, yielding superior perceptual and aesthetic qualities beyond the state-of-the-art methods that are incapable of learning from data uncertainty perspective. We hope our framework can expose a novel paradigm of data-centric learning for potential visual and language domains. Code is available.
Abstract (translated)
与现有的自然图像和视频低光增强作品相比，本研究针对动漫风景图像中的低光照质量退化问题，旨在弥合领域差异。为了应对这一尚未充分探索的增强任务，我们首先从各种来源收集图片，并构建了一个包含多样化环境和光照条件的无配对动漫风景数据集，以解决数据稀缺的问题。 为了利用与多样化的照明条件固有的不确定性信息的力量，我们提出了一种数据相对论不确定性（DRU）框架，这一框架受到了相对论生成对抗网络思想的启发。通过将光的波粒二象性类比到我们的框架中，我们明确地定义和量化了暗/亮样本的光照不确定性，并利用这种不确定性动态调整目标函数以在数据不确定性的情况下重新校准模型学习。 大量的实验展示了DRU框架的有效性，通过对几个版本的EnlightenGAN进行训练，结果超越了当前最先进的方法，在感知质量和美学质量方面均表现出色。我们希望我们的框架可以揭示一种新的以数据为中心的学习范式，适用于潜在的视觉和语言领域。代码已公开提供。
URL
https://arxiv.org/abs/2512.21944
