Update README.md
Browse files
README.md
CHANGED
|
@@ -31,6 +31,8 @@ The following table shows the performance (AP) difference between the classifier
|
|
| 31 |
| Base | 0.995 | 0.996 | 0.947 | 0.985 | 0.998 | 0.984 |
|
| 32 |
| Small | 0.986 | 0.995 | 0.888 | 0.852 | 0.993 | 0.943 |
|
| 33 |
|
|
|
|
|
|
|
| 34 |
## Dataset Summary
|
| 35 |
- The Community Forensics (small) dataset is intended for developing and benchmarking forensics methods that detect or analyze AI-generated images. It contains 278K generated images collected from 4803 generator models, and paired with 278K "real" images, sourced from [FFHQ](https://github.com/NVlabs/ffhq-dataset), [VISION](https://lesc.dinfo.unifi.it/VISION/), [COCO](https://cocodataset.org/), and [Landscapes HQ](https://github.com/universome/alis) datasets.
|
| 36 |
|
|
|
|
| 31 |
| Base | 0.995 | 0.996 | 0.947 | 0.985 | 0.998 | 0.984 |
|
| 32 |
| Small | 0.986 | 0.995 | 0.888 | 0.852 | 0.993 | 0.943 |
|
| 33 |
|
| 34 |
+
*Please check [Community Forensics-Eval](https://huggingface.co/datasets/OwensLab/CommunityForensics-Eval) for the recommended 'comprehensive' evaluation set.*
|
| 35 |
+
|
| 36 |
## Dataset Summary
|
| 37 |
- The Community Forensics (small) dataset is intended for developing and benchmarking forensics methods that detect or analyze AI-generated images. It contains 278K generated images collected from 4803 generator models, and paired with 278K "real" images, sourced from [FFHQ](https://github.com/NVlabs/ffhq-dataset), [VISION](https://lesc.dinfo.unifi.it/VISION/), [COCO](https://cocodataset.org/), and [Landscapes HQ](https://github.com/universome/alis) datasets.
|
| 38 |
|