Papers › SIDL: A Real-World Dataset for Restoring Smartphone Images with Dirty Lenses

SIDL: A Real-World Dataset for Restoring Smartphone Images with Dirty Lenses

11 Apr 2025Proceedings of the AAAI Conference on Artificial Intelligence 2025 4archive 2025-07-28

Sooyoung Choi, Sungyong Park, Heewon Kim

Smartphone cameras are ubiquitous in daily life, yet their performance can be severely impacted by dirty lenses, leading to degraded image quality. This issue is often overlooked in image restoration research, which assumes ideal or controlled lens conditions. To address this gap, we introduced SIDL (Smartphone Images with Dirty Lenses), a novel dataset designed to restore images captured through contaminated smartphone lenses. SIDL contains diverse real-world images taken under various lighting conditions and environments. These images feature a wide range of lens contaminants, including water drops, fingerprints, and dust. Each contaminated image is paired with a clean reference image, enabling supervised learning approaches for restoration tasks. To evaluate the challenge posed by SIDL, various state-of-the-art restoration models were trained and compared on this dataset. Their performances achieved some level of restoration but did not adequately address the diverse and realistic nature of the lens contaminants in SIDL. This challenge highlights the need for more robust and adaptable image restoration techniques for restoring images with dirty lenses.

PaperPDF

Code

No code repository is listed for this paper in the archive or in Syntology's graph.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Image Restoration

Datasets

Introduced by this paper, per the archive.

SIDL: A Real-World Dataset for Restoring Smartphone Images with Dirty Lenses

Results from the paper archive 2025-07-28

No leaderboard rows for this paper in the archive.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections