File size: 5,062 Bytes
e4fef82
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
#!/usr/bin/env python3
# -*- coding: utf-8 -*-
"""
Video Analysis Tab - Global analysis functions for entire videos
@author: standarduser
"""
import gradio as gr
import cv2
import numpy as np
from PIL import Image
import matplotlib.cm as cm


def compute_mean_fft(frames, progress=gr.Progress()):
    """Computes mean FFT across all video frames"""
    if not frames or len(frames) == 0:
        return None, "No video loaded"
    
    progress(0, desc="Starting Mean FFT calculation...")
    
    # Initialize accumulator for FFT magnitudes
    first_frame = frames[0]
    if len(first_frame.shape) == 3:
        gray = cv2.cvtColor(first_frame, cv2.COLOR_RGB2GRAY)
    else:
        gray = first_frame
    
    h, w = gray.shape
    fft_accumulator = np.zeros((h, w), dtype=np.float64)
    
    # Process each frame
    total_frames = len(frames)
    for i, frame in enumerate(frames):
        # Convert to grayscale
        if len(frame.shape) == 3:
            gray = cv2.cvtColor(frame, cv2.COLOR_RGB2GRAY)
        else:
            gray = frame
        
        # Compute FFT
        f_transform = np.fft.fft2(gray)
        f_shift = np.fft.fftshift(f_transform)
        
        # Magnitude spectrum (log scale)
        magnitude_spectrum = np.abs(f_shift)
        magnitude_spectrum = np.log1p(magnitude_spectrum)
        
        # Accumulate
        fft_accumulator += magnitude_spectrum
        
        # Update progress
        if i % 10 == 0:  # Update every 10 frames
            progress((i + 1) / total_frames, desc=f"Processing frame {i+1}/{total_frames}")
    
    # Compute mean
    mean_fft = fft_accumulator / total_frames
    
    progress(1.0, desc="Applying colormap...")
    
    # Normalize to 0-1
    if np.max(mean_fft) > 0:
        mean_fft_norm = mean_fft / np.max(mean_fft)
    else:
        mean_fft_norm = np.zeros_like(mean_fft)
    
    # Apply viridis colormap
    viridis = cm.get_cmap('viridis')
    colored = viridis(mean_fft_norm)
    
    # Convert to RGB and scale to 0-255
    rgb = (colored[:, :, :3] * 255).astype(np.uint8)
    result_image = Image.fromarray(rgb)
    
    status_text = f"✓ Mean FFT calculated from {total_frames} frames"
    
    return result_image, status_text


def create_tab_video_analysis(tab_label):
    """Creates a tab for video-level analysis"""
    with gr.TabItem(tab_label):
        # State for video frames (will be shared from main tab)
        video_frames = gr.State([])
        mean_fft_result = gr.State(None)
        
        gr.Markdown("# Video Analysis")
        gr.Markdown("Global analysis functions that process the entire video")
        
        with gr.Accordion("Mean FFT Analysis", open=True):
            gr.Markdown("""
            **Purpose:** Calculate the mean FFT (Fast Fourier Transform) across all video frames.
            
            This analysis helps detect:
            - Consistent frequency patterns in AI-generated videos
            - Generator-specific artifacts
            - Periodic noise or compression artifacts
            
            *Note: This operation processes all frames and may take some time for long videos.*
            """)
            
            with gr.Row():
                btn_calculate_mean_fft = gr.Button(
                    "🔬 Calculate Mean FFT",
                    variant="primary",
                    size="lg",
                    scale=1
                )
                status_text = gr.Textbox(
                    label="Status",
                    value="Ready - Load a video first",
                    interactive=False,
                    scale=2
                )
            
            with gr.Row():
                mean_fft_image = gr.Image(
                    label="Mean FFT Spectrum",
                    type="pil",
                    height=500,
                    interactive=False,    # Read-only
                    sources=[]           # No upload sources
                )
            
            with gr.Row():
                btn_download = gr.Button("💾 Download Result", size="sm")
        
        # Placeholder for future analysis tools
        with gr.Accordion("Future Analysis Tools", open=False):
            gr.Markdown("""
            **Coming soon:**
            - Temporal Consistency Analysis
            - Motion Pattern Detection
            - Frame-to-Frame Similarity Analysis
            - Audio-Video Synchronization Check
            """)
        
        # Event handlers
        btn_calculate_mean_fft.click(
            fn=compute_mean_fft,
            inputs=[video_frames],
            outputs=[mean_fft_image, status_text]
        )
        
        # Download functionality (triggers browser download)
        # Note: In Gradio, the image component already has download capability
        btn_download.click(
            fn=lambda img: img,
            inputs=[mean_fft_image],
            outputs=[mean_fft_image]
        )
        
        # Return the video_frames state so it can be connected from the main tab
        return video_frames