NTIRE 2026 challenge advances low-light image enhancement with new SOTA results

The NTIRE 2026 Low-light Enhancement: Twilight Cowboy Challenge focused on merging misaligned smartphone images captured in low-light conditions into a single, clean image. The competition addressed issues like high noise, mixed lighting, and geometric inconsistencies from hand movement. A dataset of 585 real-world scenes was used for training and benchmarking, with a three-stage evaluation protocol including blind assessment. Ten teams exceeded the baseline, achieving up to +6.49 dB in PSNR and +0.0101 in SSIM, setting new state-of-the-art results for burst-based low-light image enhancement. AI

IMPACT Establishes new benchmarks for low-light image enhancement, potentially improving smartphone photography and computational imaging.

RANK_REASON The item describes a research paper detailing a competition and its results on a specific computer vision task. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• NTIRE 2026 Low-light Enhancement: Twilight Cowboy Challenge

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →