Deckard618/imdet

AI-generated image detectors (ONNX)

Three small binary classifiers (fake / real) exported to ONNX for client-side inference with ONNX Runtime (Web or Python). In a browser, images are analyzed on the device and never uploaded.

Folder	Base model	Size	Input	Reported results
pe_tiny	Perception Encoder PE-Core T16 (Meta), 6M parameters	12.5 MB	3×256×256	94.6% balanced accuracy on DeepDetect-25 faces (92.0% after compression or blur). Catches about 65% of fakes from the newest generators.
pe_small	Perception Encoder PE-Core S16 (Meta), 24M parameters	47 MB	3×256×256	93.6% balanced accuracy on DeepDetect-25 faces. Catches about 70% of fakes from the newest generators.
resnet50-6ch	ResNet-50 with a 6-channel input	98 MB	6×224×224	97.8% accuracy (AUC 0.997) on a held-out split of its own training data.

"Newest generators" means GPT-Image-1.5, Nano Banana Pro, Seedream 4.5, Imagen 4 and FLUX.2.

Each folder holds model.onnx and a config.json that describes the preprocessing for the web app.

• Output: 2 logits, index 0 = fake, index 1 = real. Apply a softmax to get probabilities.
• pe_tiny, pe_small: resize the shorter side to 256 (bilinear), center crop 256×256, RGB scaled to [0, 1], no mean/std normalization.
• resnet50-6ch: resize to 224×224 (bicubic), then 6 channels: R, G, B / 255; HSV saturation / 255; SRM noise residual of the grayscale image + 0.5, clipped to [0, 1]; FFT log-magnitude of the grayscale image, min-max normalized to [0, 1].

Usage (Python)

import cv2
import numpy as np
import onnxruntime as ort
from PIL import Image
from scipy.ndimage import convolve
from torchvision import transforms

LABELS = ["fake", "real"]

pe_transform = transforms.Compose([
    transforms.Resize(256),
    transforms.CenterCrop(256),
    transforms.ToTensor(),
])

def pe_input(path):  # pe_tiny, pe_small
    return pe_transform(Image.open(path).convert("RGB")).unsqueeze(0).numpy()

SRM = np.array([[-1, -1, -1, -1, -1],
                [-1,  2,  2,  2, -1],
                [-1,  2,  8,  2, -1],
                [-1,  2,  2,  2, -1],
                [-1, -1, -1, -1, -1]], dtype=np.float32) / 16

def forensic6_input(path):  # resnet50-6ch
    rgb = np.asarray(Image.open(path).convert("RGB").resize((224, 224), Image.BICUBIC))
    gray = cv2.cvtColor(rgb, cv2.COLOR_RGB2GRAY).astype(np.float32) / 255
    sat = cv2.cvtColor(rgb, cv2.COLOR_RGB2HSV)[..., 1].astype(np.float32) / 255
    srm = np.clip(convolve(gray, SRM, mode="constant") + 0.5, 0, 1)
    mag = np.log1p(np.abs(np.fft.fftshift(np.fft.fft2(gray))))
    fft = (mag - mag.min()) / (mag.max() - mag.min() + 1e-8)
    x = np.concatenate([rgb.transpose(2, 0, 1) / 255, sat[None], srm[None], fft[None]])
    return x[None].astype(np.float32)

def predict(model_dir, x):
    session = ort.InferenceSession(f"{model_dir}/model.onnx")
    logits = session.run(None, {session.get_inputs()[0].name: x})[0][0]
    probs = np.exp(logits - logits.max())
    probs /= probs.sum()
    return dict(zip(LABELS, probs.tolist()))

print(predict("pe_tiny", pe_input("photo.jpg")))
print(predict("resnet50-6ch", forensic6_input("photo.jpg")))

Intended use and limitations

These models are meant for education, research and personal checks of images. A score is a probability, not proof.

• Errors happen both ways: real photos flagged as generated, and generated images missed. Recent generators are missed about 30-35% of the time by the PE models.
• resnet50-6ch is measured on a split of its own training data. Expect much lower accuracy on images from other sources.
• Compression, resizing, screenshots, filters and edits change the scores.
• Do not use a score as the only basis for a decision about a person (legal, disciplinary, professional, journalistic or other).

License

The models keep the licenses of the open models they are built on. Each license covers the model.onnx and config.json of its folder.

Folder	Built on	License	Full text
pe_tiny	Perception Encoder PE-Core T16 (Meta)	Apache 2.0	pe_tiny/LICENSE-pe_tiny.txt
pe_small	Perception Encoder PE-Core S16 (Meta)	Apache 2.0	pe_small/LICENSE-pe_small.txt
resnet50-6ch	ResNet-50 from torchvision	BSD 3-Clause	resnet50-6ch/LICENSE-resnet50-6ch.txt

Sources: Perception Encoder code and paper; torchvision code; ResNet paper. NOTICE lists the attributions and the changes made to the original models.

The models are provided as is, without any warranty.

Citation

If you use pe_tiny or pe_small, please also cite Perception Encoder:

@article{bolya2025PerceptionEncoder,
  title={Perception Encoder: The best visual embeddings are not at the output of the network},
  author={Daniel Bolya and Po-Yao Huang and Peize Sun and Jang Hyun Cho and Andrea Madotto and Chen Wei and Tengyu Ma and Jiale Zhi and Jathushan Rajasegaran and Hanoona Rasheed and Junke Wang and Marco Monteiro and Hu Xu and Shiyu Dong and Nikhila Ravi and Daniel Li and Piotr Doll{\'a}r and Christoph Feichtenhofer},
  journal={arXiv},
  year={2025}
}

If you use resnet50-6ch, please also cite ResNet:

@inproceedings{he2016deep,
  title={Deep Residual Learning for Image Recognition},
  author={He, Kaiming and Zhang, Xiangyu and Ren, Shaoqing and Sun, Jian},
  booktitle={Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  pages={770--778},
  year={2016}
}