File size: 2,121 Bytes
3a3c2be
cb82c24
3a3c2be
 
 
 
 
 
 
 
cb82c24
 
 
 
3a3c2be
cb82c24
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
3a3c2be
cb82c24
 
 
 
 
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
47
48
49
50
51
52
53
54
55
56
57
import os
import gradio as gr

# Clone the GitHub repository containing app.py if not already cloned
REPO_URL = "https://github.com/NeeravSood/AllMark-MVP"  # Replace with your GitHub repo URL
if not os.path.exists("AllMark-MVP"):  # Replace with your repo's folder name
    os.system(f"git clone {REPO_URL}")

# Import the backend code after cloning the repo
from repository_name.app import DeepfakeAnalyzer  # Adjust based on your repo and file structure

# Initialize the analyzer instance
analyzer = DeepfakeAnalyzer()

# Define the function for Gradio to call
def analyze_video(video_file):
    results = analyzer.analyze_media(video_file)
    combined_probability = results['combined_assessment']
    audio_analysis = results["audio_analysis"]
    video_probability = results['video_analysis']['probability']
    frame_count = len(results['video_analysis']['frame_results'])

    # Format the output for display
    output = {
        "Audio Analysis": audio_analysis,
        "Deepfake Probability (Combined Assessment)": f"{combined_probability:.2f}%",
        "Video Analysis": {
            "Deepfake Probability": f"{video_probability:.4f}",
            "Frames Analyzed": frame_count,
            "Frame Analysis Summary": [
                {
                    "Frame Number": frame_result["frame_number"],
                    "Noise Level": frame_result["noise"],
                    "Edge Density": frame_result["edge_density"],
                    "Color Consistency": frame_result["color_consistency"],
                    "Temporal Difference": frame_result["temporal_difference"],
                    "Probability": frame_result["probability"]
                }
                for frame_result in results['video_analysis']['frame_results']
            ]
        }
    }
    return output

# Define the Gradio interface
interface = gr.Interface(
    fn=analyze_video,
    inputs=gr.Video(label="Upload Video"),
    outputs="json",
    title="Deepfake Analyzer",
    description="Upload a video to analyze for deepfake content."
)

# Launch Gradio app
if __name__ == "__main__":
    interface.launch()