Anomaly Detection on Imagenette + Ambivision (test)
0.9985AUC-ROCIsolation Forest
Evaluation results
| Method | Links | |
|---|---|---|
| Isolation Forest2026.03 | 0.9985 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=in (after GELU and Convolution-Layers), Quantile=0.0012026.03 | 0.9945 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=after (after Transformer-Block), Quantile=0.0012026.03 | 0.9937 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=after (after Transformer-Block), Quantile=0.012026.03 | 0.9923 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=in (after GELU and Convolution-Layers), Quantile=0.012026.03 | 0.9893 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=in (after GELU and Convolution-Layers), Quantile=0.012026.03 | 0.971 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=in (after GELU and Convolution-Layers), Quantile=0.0012026.03 | 0.9703 | |
| AutoencoderTraining data=10% of the INet testset2026.03 | 0.9472 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=after (after Transformer-Block), Quantile=0.0012026.03 | 0.9357 | |
| RangeADBackbone=Swin-v2-Vision Transformer [26], Feature extraction point=after (after Transformer-Block), Quantile=0.012026.03 | 0.9355 | |
| DeepSVDDTraining data=10% of the INet testset2026.03 | 0.883 | |
| DeepSVDDTraining data=10% of the INet testset2026.03 | 0.6833 | |
| Isolation Forest2026.03 | 0.6597 | |
| AutoencoderTraining data=10% of the INet testset2026.03 | 0.4266 |