Consists of 8,422 blurry and sharp image pairs with 65,784 densely annotated FG human bounding boxes.
Source: Human-Aware Motion Deblurring
Variants: HIDE, HIDE (trained on GOPRO)
This dataset is used in 2 benchmarks:
| Task | Model | Paper | Date |
|---|---|---|---|
| Image Deblurring | AdaRevD | AdaRevD: Adaptive Patch Exiting Reversible … | 2024-06-13 |
| Image Deblurring | EVSSM | Efficient Visual State Space Model … | 2024-05-23 |
| Image Deblurring | ALGNet-B | Learning Enriched Features via Selective … | 2024-03-29 |
| Image Deblurring | M3SNet | A Mountain-Shaped Single-Stage Network for … | 2023-05-09 |
| Deblurring | MAXIM-3S | MAXIM: Multi-Axis MLP for Image … | 2022-01-09 |
| Image Deblurring | MAXIM-3S | MAXIM: Multi-Axis MLP for Image … | 2022-01-09 |
Recent papers with results on this dataset: