# Exposure Correction on Combined over and underexposed images 1.0 (test) Source: [Exposure Correction on Combined over and underexposed images 1.0 (test)]() Collection index: [llms.txt]() Data updated: 2026-02-26T05:57:05.792Z Evaluation of Exposure Correction performance on the test split of the 'Combined over and underexposed images 1.0' dataset, utilizing image quality metrics (PSNR and SSIM) assessed independently by five different experts (A through E), along with calculated averages and the PI metric. A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. | Scope | Recorded value | | --- | --- | | Task | Exposure Correction | | Dataset (including any stated split) | Combined over and underexposed images 1.0 (test) | | Metric | PSNR (Expert A) | | Unit | SCALAR | | Direction | Higher is better | | Results loaded | 18 of 18 | | Coverage | All results currently returned for this benchmark | Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol. ## Metrics - [PSNR (Expert A)](): Higher is better; unit: SCALAR. - [SSIM (Expert A)](): Higher is better; unit: SCALAR. - [PSNR (Expert B)](): Higher is better; unit: SCALAR. - [SSIM (Expert B)](): Higher is better; unit: SCALAR. - [PSNR (Expert C)](): Higher is better; unit: SCALAR. - [SSIM (Expert C)](): Higher is better; unit: SCALAR. - [PSNR (Expert D)](): Higher is better; unit: SCALAR. - [SSIM (Expert D)](): Higher is better; unit: SCALAR. - [PSNR (Expert E)](): Higher is better; unit: SCALAR. - [SSIM (Expert E)](): Higher is better; unit: SCALAR. - [Average PSNR](): Higher is better; unit: SCALAR. - [Average SSIM](): Higher is better; unit: SCALAR. - [PI](): Higher is better; unit: SCALAR. ## Recorded results | Method | PSNR (Expert A) | Result date | Report paper | | --- | --- | --- | --- | | coarse-to-fine DNN | 19.158 | Not recorded | 68fa69981e209131cea22ee4 | | coarse-to-fine DNN | 19.114 | Not recorded | 68fa69981e209131cea22ee4 | | DPED | 18.45 | Not recorded | 68fa69981e209131cea22ee4 | | DPE | 16.933 | Not recorded | 68fa69981e209131cea22ee4 | | DPED | 16.91 | Not recorded | 68fa69981e209131cea22ee4 | | DPE | 16.24 | Not recorded | 68fa69981e209131cea22ee4 | | HE | 16.148 | Not recorded | 68fa69981e209131cea22ee4 | | HDR CNN | 15.812 | Not recorded | 68fa69981e209131cea22ee4 | | DPE | 15.69 | Not recorded | 68fa69981e209131cea22ee4 | | DPED | 15.134 | Not recorded | 68fa69981e209131cea22ee4 | | CLAHE | 14.884 | Not recorded | 68fa69981e209131cea22ee4 | | WVM | 14.488 | Not recorded | 68fa69981e209131cea22ee4 | | HDR CNN | 13.709 | Not recorded | 68fa69981e209131cea22ee4 | | HQEC | 13.385 | Not recorded | 68fa69981e209131cea22ee4 | | Deep UPE | 13.161 | Not recorded | 68fa69981e209131cea22ee4 | | Zero-DCE | 11.643 | Not recorded | 68fa69981e209131cea22ee4 | | LIME | 11.154 | Not recorded | 68fa69981e209131cea22ee4 | | RetinexNet | 10.759 | Not recorded | 68fa69981e209131cea22ee4 | ### coarse-to-fine DNN PSNR (Expert A): 19.158. Result ID: 698dbe098b517eb433237329. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Adversarial Loss":"w/o Ladv"}. This record does not assert independent reproduction. ### coarse-to-fine DNN PSNR (Expert A): 19.114. Result ID: 698dbe098b517eb43323732a. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Adversarial Loss":"w/ Ladv"}. This record does not assert independent reproduction. ### DPED PSNR (Expert A): 18.45. Result ID: 698dbe098b517eb433237321. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Backbone":"Sony"}. This record does not assert independent reproduction. ### DPE PSNR (Expert A): 16.933. Result ID: 698dbe098b517eb433237324. [Reporting paper]() Recorded evaluation configuration: {"Variant":"S-FiveK","Supervision":"Supervised"}. This record does not assert independent reproduction. ### DPED PSNR (Expert A): 16.91. Result ID: 698dbe098b517eb433237320. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Backbone":"BlackBerry"}. This record does not assert independent reproduction. ### DPE PSNR (Expert A): 16.24. Result ID: 698dbe098b517eb433237323. [Reporting paper]() Recorded evaluation configuration: {"Variant":"U-FiveK","Supervision":"Unsupervised"}. This record does not assert independent reproduction. ### HE PSNR (Expert A): 16.148. Result ID: 698dbe098b517eb433237319. [Reporting paper]() Recorded evaluation configuration: {"non-deep learning":true}. This record does not assert independent reproduction. ### HDR CNN PSNR (Expert A): 15.812. Result ID: 698dbe098b517eb43323731e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Refining Module":"PS"}. This record does not assert independent reproduction. ### DPE PSNR (Expert A): 15.69. Result ID: 698dbe098b517eb433237322. [Reporting paper]() Recorded evaluation configuration: {"Variant":"HDR"}. This record does not assert independent reproduction. ### DPED PSNR (Expert A): 15.134. Result ID: 698dbe098b517eb43323731f. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"Model Backbone":"iPhone"}. This record does not assert independent reproduction. ### CLAHE PSNR (Expert A): 14.884. Result ID: 698dbe098b517eb43323731a. [Reporting paper]() Recorded evaluation configuration: {"non-deep learning":true}. This record does not assert independent reproduction. ### WVM PSNR (Expert A): 14.488. Result ID: 698dbe098b517eb43323731b. [Reporting paper]() Recorded evaluation configuration: {"non-deep learning":true}. This record does not assert independent reproduction. ### HDR CNN PSNR (Expert A): 13.709. Result ID: 698dbe098b517eb43323731d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Refining Module":"RHT"}. This record does not assert independent reproduction. ### HQEC PSNR (Expert A): 13.385. Result ID: 698dbe098b517eb433237325. [Reporting paper]() Recorded evaluation configuration: {"non-deep learning":true}. This record does not assert independent reproduction. ### Deep UPE PSNR (Expert A): 13.161. Result ID: 698dbe098b517eb433237327. [Reporting paper]() Recorded evaluation configuration: {"optimized for":"underexposure correction"}. This record does not assert independent reproduction. ### Zero-DCE PSNR (Expert A): 11.643. Result ID: 698dbe098b517eb433237328. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"optimized for":"underexposure correction"}. This record does not assert independent reproduction. ### LIME PSNR (Expert A): 11.154. Result ID: 698dbe098b517eb43323731c. [Reporting paper]() Recorded evaluation configuration: {"non-deep learning":true,"optimized for":"underexposure correction"}. This record does not assert independent reproduction. ### RetinexNet PSNR (Expert A): 10.759. Result ID: 698dbe098b517eb433237326. [Reporting paper]() [Method origin paper]() Recorded evaluation configuration: {"optimized for":"underexposure correction"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.