AIM-500, short for Automatic Image Matting-500, is the first natural image matting test set, which contains 500 high-resolution real-world natural images. The images come from three types (salient opaque foreground, significant transparent/clean foreground, non-salient foreground), and multiple categories.

Citation

@inproceedings{ijcai2021-111, title = {Deep Automatic Natural Image Matting}, author = {Li, Jizhizi and Zhang, Jing and Tao, Dacheng} booktitle = {Proceedings of the Thirtieth International Joint Conference on Artificial Intelligence, {IJCAI-21}}, publisher = {International Joint Conferences on Artificial Intelligence Organization}, editor = {Zhi-Hua Zhou}, pages = {800–806}, year = {2021}, month = {8}, note = {Main Track} doi = {10.24963/ijcai.2021/111}, url = {https://doi.org/10.24963/ijcai.2021/111}, }