Konstanz artificially distorted image quality database (KADID-10k) contains 81 pristine images, each degraded by 25 distortions in 5 levels.
Variants: KADID-10k
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| No-Reference Image Quality Assessment | ARNIQA | ARNIQA: Learning Distortion Manifold for … | 2023-10-20 |
| No-Reference Image Quality Assessment | UNIQA | You Only Train Once: A … | 2023-10-14 |
| No-Reference Image Quality Assessment | Re-IQA | Re-IQA: Unsupervised Learning for Image … | 2023-04-02 |
| No-Reference Image Quality Assessment | CONTRIQUE | Image Quality Assessment using Contrastive … | 2021-10-25 |
| No-Reference Image Quality Assessment | TReS | No-Reference Image Quality Assessment via … | 2021-08-16 |
| No-Reference Image Quality Assessment | DB-CNN | Blind Image Quality Assessment Using … | 2019-07-05 |
Recent papers with results on this dataset: