lukhsaankumar/DeepFakeDetectorBackend-storage / .hf_cache /DeepFakeDetector--cnn-transfer-final /README.md
| license: apache-2.0 | |
| tags: | |
| - image-classification | |
| - deepfake-detection | |
| - pytorch | |
| # CNN Transfer (DeepFakeDetector) | |
| Binary image classifier for deepfake or AI-generated image detection. | |
| ## Label convention (fixed) | |
| - 0 = real | |
| - 1 = fake | |
| ## Preprocessing | |
| - Resize to 224x224 | |
| - Normalize with ImageNet mean/std | |
| ## Output | |
| - prob_fake in [0, 1] computed as softmax(logits)[fake] | |
| - decision threshold: 0.024 | |
Xet Storage Details
- Size:
- 419 Bytes
- Xet hash:
- 7c4fe78fd034a6c54d38c9ef35711904b5d53e15ae8b153d13381036cb46bb9c
·
Xet efficiently stores files, intelligently splitting them into unique chunks and accelerating uploads and downloads. More info.