Automatic Image Matting-500
AIM-500 is the first natural image matting test set, contains 500 high-resolution real-world natural images from three types of images (salient opaque foregrounds, salient transparent/meticulous foregrounds, non-salient foregrounds), and multiple categories. The amount of each category is shown in the following table.
| Portrait | Animal | Transparent | Plant | Furniture | Toy | Fruit |
|---|---|---|---|---|---|---|
| 100 | 200 | 34 | 75 | 45 | 36 | 10 |
Variants: AIM-500
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Image Matting | DiffMatte | Diffusion for Natural Image Matting | 2023-12-10 |
| Image Matting | AIM-Net | Deep Automatic Natural Image Matting | 2021-07-15 |
| Image Matting | GFM | Bridging Composite and Real: Towards … | 2020-10-30 |
| Image Matting | U2NET | U2-Net: A Bayesian U-Net model … | 2019-01-23 |
| Image Matting | SHM | Semantic Human Matting | 2018-09-05 |
Recent papers with results on this dataset: