Spaces:
Sleeping
Sleeping
File size: 1,632 Bytes
e73e2a2 3a2cd79 e73e2a2 3a2cd79 e6fe7fe 3a2cd79 e73e2a2 eaa1a24 ce59aac e73e2a2 3a2cd79 ce59aac 93ce865 e6fe7fe ce59aac e73e2a2 93ce865 ce59aac 93ce865 4ab3d47 3a2cd79 ce59aac 3a2cd79 ce59aac e73e2a2 3a2cd79 777870a ce59aac 2bf05af e73e2a2 9edd18a |
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 |
import os
import cv2
import numpy as np
import gradio as gr
import tensorflow as tf
from tensorflow.keras.models import load_model
from tensorflow.keras.applications.xception import preprocess_input as xcp_pre
from tensorflow.keras.applications.efficientnet import preprocess_input as eff_pre
from huggingface_hub import hf_hub_download
# Load models
xcp_path = hf_hub_download(repo_id="Zeyadd-Mostaffa/deepfake-image-detector", filename="xception_model.h5")
eff_path = hf_hub_download(repo_id="Zeyadd-Mostaffa/deepfake-image-detector", filename="efficientnet_model.h5")
xcp_model = load_model(xcp_path)
eff_model = load_model(eff_path)
def predict(image):
# Resize for each model
xcp_img = cv2.resize(image, (299, 299))
eff_img = cv2.resize(image, (224, 224))
# Preprocess
xcp_tensor = xcp_pre(xcp_img.astype(np.float32))[np.newaxis, ...]
eff_tensor = eff_pre(eff_img.astype(np.float32))[np.newaxis, ...]
# Predict
xcp_pred = xcp_model.predict(xcp_tensor, verbose=0).flatten()[0]
eff_pred = eff_model.predict(eff_tensor, verbose=0).flatten()[0]
avg_pred = (xcp_pred + eff_pred) / 2
label = "Real" if avg_pred > 0.5 else "Fake"
# ✅ Simplest format: just the label (for gradio_client compatibility)
return label
# ✅ Output changed from gr.Label → gr.Textbox to avoid JSON schema issues
interface = gr.Interface(
fn=predict,
inputs=gr.Image(type="numpy", label="Upload Image"),
outputs=gr.Textbox(label="Prediction"),
title="Deepfake Image Detector",
description="Upload a full image. The model classifies it as real or fake."
)
interface.launch()
|