{"ID":2832563,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.08982","arxiv_id":"2512.08982","title":"Consist-Retinex: One-Step Noise-Emphasized Consistency Training Accelerates High-Quality Retinex Enhancement","abstract":"Retinex-based low-light image enhancement benefits from separating reflectance and illumination, yet recent generative approaches often rely on iterative sampling and are difficult to deploy under strict latency budgets. Consistency models offer a natural route to one-step restoration, but direct adaptation to Retinex-factorized enhancement is unstable: one-step inference is evaluated at the high-noise endpoint, whereas standard training schedules provide little supervision there, and temporal self-consistency alone does not determine the correct conditional target. We propose Consist-Retinex, which first uses a Retinex Transformer Decomposition Network (TDN) to obtain paired reflectance and illumination maps, then trains two conditional consistency models with a Retinex-aware dual objective and adaptive noise-emphasized fixed-point sampling. The dual objective combines trajectory consistency with paired ground-truth component alignment, while the sampling rule concentrates supervision near the inference endpoint without discarding full-range noise coverage. We further provide an endpoint error bound, an anchoring-propagation result, and a high-noise sample-allocation analysis that explain why endpoint supervision and temporal consistency are complementary for one-step Retinex enhancement. Experiments on paired and unpaired low-light benchmarks show that Consist-Retinex obtains the best VE-LOL-L scores among the compared methods under one-step inference and remains competitive on LOL, with substantially reduced sampling and consistency-stage training cost in the reported setup.","short_abstract":"Retinex-based low-light image enhancement benefits from separating reflectance and illumination, yet recent generative approaches often rely on iterative sampling and are difficult to deploy under strict latency budgets. Consistency models offer a natural route to one-step restoration, but direct adaptation to Retinex-...","url_abs":"https://arxiv.org/abs/2512.08982","url_pdf":"https://arxiv.org/pdf/2512.08982v3","authors":"[\"Jian Xu\",\"Wei Chen\",\"Shigui Li\",\"Delu Zeng\",\"John Paisley\",\"Qibin Zhao\"]","published":"2025-12-05T13:44:19Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Transformer\"]","has_code":false}