ADE-OoD is a public benchmark for dense out-of-distribution detection in general natural images. It measures the ability to detect and localize objects which are out-of-distribution with respect to the 150 categories of the ADE20k semantic segmentation dataset.
The benchmark data and annotations are available for download at the project page: ade-ood.github.io/
Variants: ADE-OoD
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Out-of-Distribution Detection | DOoD | Diffusion for Out-of-Distribution Detection on … | 2024-07-22 |
| Out-of-Distribution Detection | RbA | RbA: Segmenting Unknown Regions Rejected … | 2022-11-25 |
| Out-of-Distribution Detection | cDNP | Far Away in the Deep … | 2022-11-12 |
| Out-of-Distribution Detection | GMMSeg | GMMSeg: Gaussian Mixture based Generative … | 2022-10-05 |
Recent papers with results on this dataset: