# Semantic Segmentation on ADE20K-C 15 corruptions (test) Source: [Semantic Segmentation on ADE20K-C 15 corruptions (test)]() Collection index: [llms.txt]() Data updated: 2026-04-02T08:42:47.937Z Evaluation of Semantic Segmentation robustness on the test split of the ADE20K-C dataset, measuring performance across 15 specific image corruption types (e.g., brightness, fog, JPEG compression) and their average. 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 | Semantic Segmentation | | Dataset (including any stated split) | ADE20K-C 15 corruptions (test) | | Metric | Brightness | | Unit | PERCENT | | Direction | Higher is better | | Results loaded | 8 of 8 | | 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 - [Brightness](): Higher is better; unit: PERCENT. - [Contrast](): Higher is better; unit: PERCENT. - [Defocus](): Higher is better; unit: PERCENT. - [Elastic Deformation](): Higher is better; unit: PERCENT. - [Fog](): Higher is better; unit: PERCENT. - [Frost](): Higher is better; unit: PERCENT. - [Gaussian Noise](): Higher is better; unit: PERCENT. - [Glass Blur](): Higher is better; unit: PERCENT. - [Impulse Noise](): Higher is better; unit: PERCENT. - [JPEG Compression](): Higher is better; unit: PERCENT. - [Motion Blur](): Higher is better; unit: PERCENT. - [Pixelate](): Higher is better; unit: PERCENT. - [Shot Noise](): Higher is better; unit: PERCENT. - [Snow](): Higher is better; unit: PERCENT. - [Zoom Blur](): Higher is better; unit: PERCENT. - [Average Score](): Higher is better; unit: PERCENT. - [mIoU](): Higher is better; unit: PERCENT. ## Recorded results | Method | Brightness | Result date | Report paper | | --- | --- | --- | --- | | VICT | 44.4 | Not recorded | 68fa779d1e209131cea745cd | | Painter | 44.1 | Not recorded | 68fa779d1e209131cea745cd | | VICT | 44 | Not recorded | 68fa779d1e209131cea745cd | | Painter | 43.4 | Not recorded | 68fa779d1e209131cea745cd | ### VICT Brightness: 44.4. Result ID: 699553463badfaffe643617e. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation protocol":"one-shot setting","Corruption level":"4"}. This record does not assert independent reproduction. ### Painter Brightness: 44.1. Result ID: 699553463badfaffe643617d. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation protocol":"one-shot setting","Corruption level":"4"}. This record does not assert independent reproduction. ### VICT Brightness: 44. Result ID: 699553463badfaffe643617c. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation protocol":"zero-shot setting","Corruption level":"4"}. This record does not assert independent reproduction. ### Painter Brightness: 43.4. Result ID: 699553463badfaffe643617b. [Reporting paper]() [Method origin paper]() [Implementation]() Recorded evaluation configuration: {"Evaluation protocol":"zero-shot setting","Corruption level":"4"}. This record does not assert independent reproduction. Page 1 of 1. This is an index page, not the full collection.