ControlLight: 邁向可控、一致且可泛化的低光增強

摘要

現有的深度學習低光照增強方法通常僅在有限的數據集上訓練，且針對單一增強目標，這限制了其在真實場景中的泛化能力與可控性。為克服這些限制，我們提出ControlLight——一個可控、一致且具備泛化能力的低光照增強框架。首先，我們構建了一個大規模的真實退化影像數據集，並提供連續光照強度的監控資訊。為確保在不同控制強度下輸出的表現一致性，我們引入一種對齊感知加權流匹配損失，能在連續增強強度下保留影像結構。ControlLight允許使用者透過靈活控制強度來編輯真實場景中的退化低光照影像，以達到滿意的增強結果，同時保持視覺一致性和真實感。大量實驗顯示，與現有低光照增強方法相比，ControlLight達到了最先進的效能，同時展現出強大的連續可控性以及對真實場景的泛化能力。

English

Existing deep learning-based low-light enhancement methods are typically trained on limited datasets with single enhancement targets, which restricts their generalization ability and controllability in real-world applications. To overcome these limitations, we propose ControlLight, a controllable, consistent, and generalizable framework for low-light enhancement. We first construct a large-scale dataset of real-world degraded images with continuous illumination-strength supervision. To further ensure consistent outputs under different control strengths, we introduce a misalignment-aware weighted flow matching loss that preserves image structure across continuous enhancement strengths. ControlLight allows users to edit real-world degraded low-light images toward satisfactory enhancement results by flexibly controlling the strength while preserving visual consistency and realism. Extensive experiments show that ControlLight achieves state-of-the-art performance against existing low-light enhancement approaches while demonstrating strong continuous controllability and generalization to real-world scenarios.