Composition-1K is a large-scale image matting dataset including 49300 training images and 1000 testing images.
Image source: https://arxiv.org/pdf/1703.03872v3.pdf
Variants: Composition-1K
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Image Matting | DiffMatte | Diffusion for Natural Image Matting | 2023-12-10 |
| Image Matting | ViTMatte | ViTMatte: Boosting Image Matting with … | 2023-05-24 |
| Image Matting | AEMatter | Revisiting Context Aggregation for Image … | 2023-04-03 |
| Image Matting | PP-Matting | PP-Matting: High-Accuracy Natural Image Matting | 2022-04-20 |
| Image Matting | MatteFormer | MatteFormer: Transformer-Based Image Matting via … | 2022-03-29 |
| Image Matting | LSAMatting | Improving Deep Image Matting via … | 2021-12-27 |
| Image Matting | TMFNet | Trimap-guided Feature Mining and Fusion … | 2021-12-01 |
| Image Matting | LFPNet | Long-Range Feature Propagating for Natural … | 2021-09-25 |
| Image Matting | SIM | Semantic Image Matting | 2021-04-16 |
| Image Matting | FBAMatting | $F$, $B$, Alpha Matting | 2020-03-17 |
| Image Matting | Context-Aware Matting | Context-Aware Image Matting for Simultaneous … | 2019-09-20 |
| Image Matting | IndexNet-Matting | Indices Matter: Learning to Index … | 2019-08-02 |
| Image Matting | DIM | Deep Image Matting | 2017-03-10 |
Recent papers with results on this dataset: