File size: 11,755 Bytes
adf4e19
 
 
 
 
 
 
 
 
 
59f824e
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
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
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
---

title: Deception Analysis API
emoji: πŸ•΅οΈ
colorFrom: indigo
colorTo: blue
sdk: docker
app_port: 7860
pinned: false
---


# πŸ•΅οΈβ€β™‚οΈ Explainable Multimodal Deception Analysis System

A comprehensive system for detecting deception in video interviews using facial micro-expressions, audio speech patterns, temporal dynamics, and SHAP-based explainability. Works with or without real datasets using synthetic data generation.

## πŸš€ Quick Start (Google Colab - No Local Installation)

**Run everything in the cloud with one click:**

[![Open In Colab](https://colab.research.google.com/assets/colab-badge.svg)](https://colab.research.google.com/github/apexprit/deception-analysis/blob/main/colab_deception_analysis.ipynb)

### Steps to Run in Google Colab:

1. **Open the Colab notebook** using the link above
2. **Click "Runtime" β†’ "Run all"** (Ctrl+F9)
3. **Wait for installation** (~2-3 minutes)
4. **Use the Gradio interface** that appears at the bottom

> [!TIP]
> **Working locally?** Make sure to `git push` your latest changes to GitHub before running in Colab, as the notebook clones the repository directly from the `main` branch.

The notebook will:
- βœ… Install all dependencies automatically
- βœ… Generate realistic synthetic deception data
- βœ… Train a deception detection model (~85% accuracy)
- βœ… Launch interactive web interface
- βœ… Provide sample analysis with visualizations

## πŸ“‹ Features

### πŸ” **Multimodal Analysis**
- **Facial Features**: 16 micro-expressions (eye openness, mouth movement, brow tension, head dynamics)
- **Audio Features**: 30+ speech patterns (pitch variation, pauses, MFCC coefficients, spectral features)
- **Cross-modal Features**: Audio-visual sync, pause-gaze coupling, stress composite
- **Temporal Analysis**: Frame-by-frame probability tracking, spike detection, suspicious intervals

### 🧠 **Explainability**
- **SHAP-based feature importance** with behavioral interpretations
- **Feature group analysis** (facial vs. audio vs. cross-modal)
- **Human-readable explanations** of deception indicators
- **Publication-quality visualizations**

### 🎯 **Subject Adaptation**
- **Personal calibration** to reduce inter-subject variability
- **Baseline establishment** for individual behavioral patterns
- **Adaptive thresholding** based on subject history

### πŸ“Š **Evaluation & Validation**
- **Comprehensive metrics**: Accuracy, Precision, Recall, F1, AUC-ROC
- **Cross-validation**: Stratified, subject-aware, leave-one-subject-out
- **Ablation studies**: Feature group importance analysis
- **Bootstrap confidence intervals** for robust metrics

## πŸ—οΈ Architecture

```

Video Input

    β”‚

    β”œβ”€β”€ Facial Feature Extraction (MediaPipe Face Mesh)

    β”‚   └── 16 micro-expression features

    β”‚

    β”œβ”€β”€ Audio Feature Extraction (librosa)

    β”‚   └── 30+ speech pattern features

    β”‚

    └── Multimodal Fusion

        └── Temporal alignment + cross-modal features

            β”‚

            β”œβ”€β”€ Deception Classifier (Gradient Boosting)

            β”‚   └── Probability prediction (0-1)

            β”‚

            β”œβ”€β”€ Subject Calibration

            β”‚   └── Personalized baseline adjustment

            β”‚

            β”œβ”€β”€ Temporal Analysis

            β”‚   └── Spike detection, interval localization

            β”‚

            └── Explainability Engine (SHAP)

                └── Feature importance + behavioral interpretation

```

## πŸ“ Project Structure

```

deception-analysis/

β”œβ”€β”€ app.py                          # Gradio web interface

β”œβ”€β”€ train.py                        # CLI training script

β”œβ”€β”€ inference.py                    # CLI inference script

β”œβ”€β”€ create_pretrained_model.py      # Generate pre-trained model

β”œβ”€β”€ requirements.txt                # Dependencies

β”œβ”€β”€ colab_deception_analysis.ipynb  # Google Colab notebook

β”œβ”€β”€ notebooks/

β”‚   β”œβ”€β”€ demo.ipynb                  # Full workflow demonstration

β”‚   └── demo_with_synthetic_data.ipynb  # Synthetic data demo

β”œβ”€β”€ data/                           # Generated synthetic data

β”œβ”€β”€ models/                         # Pre-trained models

└── src/                            # Core modules

    β”œβ”€β”€ pipeline.py                 # Main orchestration

    β”œβ”€β”€ facial/                     # Facial analysis (extractor, visualizer)

    β”œβ”€β”€ audio/                      # Audio analysis (extractor, fusion)

    β”œβ”€β”€ temporal/                   # Temporal analysis (analyzer, visualizer)

    β”œβ”€β”€ model/                      # Classifier + calibration

    β”œβ”€β”€ explainability/             # SHAP explanations (explainer, visualizer)

    β”œβ”€β”€ evaluation/                 # Metrics, CV, ablation, plots

    └── utils/                      # Config, I/O, synthetic data, downloader

```

## πŸ› οΈ Local Installation (Optional)

If you want to run locally:

```bash

# Clone repository

git clone https://github.com/apexprit/deception-analysis.git

cd deception-analysis



# Install dependencies

pip install -r requirements.txt



# Generate synthetic data and pre-trained model

python create_pretrained_model.py



# Train on synthetic data

python train.py --data_dir ./data/synthetic --labels_file ./data/synthetic/synthetic_labels.csv



# Run inference (demo mode)

python inference.py --model ./models/deception_model.pkl --video_dir ./data/sample/



# Launch web interface

python app.py --port 7860

```

## πŸ“Š Usage Examples

### 1. **Basic Analysis**
```python

from src.pipeline import DeceptionPipeline

from src.utils.config import AppConfig



# Initialize pipeline

config = AppConfig()

pipeline = DeceptionPipeline(config)



# Analyze a video

result = pipeline.analyze_video(

    video_path="interview.mp4",

    subject_id="S001",

    generate_visualizations=True,

    output_dir="./results"

)



print(f"Deception Probability: {result['deception_probability']:.3f}")

print(f"Prediction: {result['prediction']}")

print(f"Confidence: {result['confidence']:.3f}")

```

### 2. **Batch Processing**
```python

# Analyze multiple videos

results = pipeline.analyze_batch(

    video_paths=["video1.mp4", "video2.mp4", "video3.mp4"],

    subject_ids=["S001", "S002", "S003"],

    generate_visualizations=False,

    output_dir="./batch_results"

)

```

### 3. **Model Training**
```python

# Train on your dataset

train_result = pipeline.train(

    video_paths=["train1.mp4", "train2.mp4", "train3.mp4"],

    labels=[0, 1, 0],  # 0=truthful, 1=deceptive

    subject_ids=["S001", "S002", "S003"],

    output_dir="./models"

)



print(f"Training Accuracy: {train_result['accuracy']:.3f}")

print(f"Training F1: {train_result['f1']:.3f}")

```

### 4. **Synthetic Data Generation**
```python

from src.utils.synthetic_data import SyntheticDataGenerator



# Generate synthetic dataset

generator = SyntheticDataGenerator()

features, labels, subject_ids = generator.generate_dataset(

    n_truthful=500,

    n_deceptive=500

)



# Save for training

generator.save_synthetic_dataset(

    output_dir="./data/synthetic",

    n_truthful=500,

    n_deceptive=500,

    dataset_name="my_dataset"

)

```

## πŸ§ͺ Evaluation

```python

from src.evaluation.metrics import DeceptionMetrics

from src.evaluation.cross_validation import CrossValidator

from src.evaluation.ablation import AblationStudy



# Compute metrics

metrics = DeceptionMetrics()

all_metrics = metrics.compute_all_metrics(y_true, y_pred, y_prob)



# Cross-validation

cv = CrossValidator()

cv_results = cv.run_cross_validation(

    classifier=DeceptionClassifier,

    X=features,

    y=labels,

    n_splits=5,

    method='stratified'

)



# Ablation study

ablation = AblationStudy(feature_names, feature_groups)

ablation_results = ablation.run_ablation(

    classifier_class=DeceptionClassifier,

    X=features,

    y=labels,

    feature_names=feature_names,

    feature_groups=feature_groups,

    n_splits=5

)

```

## πŸ”§ Configuration

The system is highly configurable via `src/utils/config.py`:

```python

from src.utils.config import AppConfig



config = AppConfig(

    project=ProjectConfig(

        data_dir="./data",

        model_dir="./models",

        results_dir="./results",

        random_seed=42

    ),

    feature=FeatureConfig(

        fps=15,

        face_detection_confidence=0.5,

        landmark_refinement=True

    ),

    audio=AudioConfig(

        sample_rate=16000,

        n_fft=2048,

        hop_length=512,

        n_mfcc=13

    ),

    model=ModelConfig(

        model_type="gradient_boosting",

        n_estimators=100,

        learning_rate=0.1,

        max_depth=3

    ),

    temporal=TemporalConfig(

        spike_threshold=0.7,

        min_spike_duration=0.5,

        smoothing_window=5

    )

)

```

## πŸ“ˆ Performance

With synthetic data (1000 samples, 80/20 train/test split):
- **Accuracy**: ~85%
- **F1-Score**: ~0.83
- **AUC-ROC**: ~0.91
- **Precision**: ~0.82
- **Recall**: ~0.84

**Feature Importance** (top 5):
1. `pause_count` (audio) - More pauses indicate deception
2. `eye_left_openness` (facial) - Less eye openness indicates deception
3. `pitch_variation_coefficient` (audio) - Higher variation indicates deception
4. `brow_tension` (facial) - Higher tension indicates deception
5. `audio_visual_sync` (cross-modal) - Less sync indicates deception

## πŸš€ Deployment Options

### 1. **Google Colab** (Recommended for no local setup)
- Open `colab_deception_analysis.ipynb` in Colab
- Run all cells
- Use the Gradio interface

### 2. **Local Python Environment**
```bash

pip install -r requirements.txt

python app.py

```

### 3. **Docker Container**
```dockerfile

FROM python:3.9

COPY . /app

WORKDIR /app

RUN pip install -r requirements.txt

CMD ["python", "app.py", "--host", "0.0.0.0", "--port", "7860"]

```

### 4. **Cloud Services**
- **Hugging Face Spaces**: Deploy Gradio app
- **Vercel/Heroku**: Deploy as web service
- **AWS SageMaker**: Deploy as ML endpoint
- **Google Cloud Run**: Containerized deployment

## πŸ“š Research Basis

The system implements features based on established deception detection research:

1. **Facial Indicators** (Ekman, 2009):
   - Micro-expressions (brief facial expressions)
   - Eye gaze avoidance
   - Facial asymmetry

2. **Vocal Indicators** (DePaulo et al., 2003):
   - Higher pitch variation
   - More speech errors
   - Longer response latency

3. **Multimodal Integration** (PΓ©rez-Rosas & Mihalcea, 2014):
   - Audio-visual feature fusion
   - Temporal alignment
   - Cross-modal correlation analysis

## ⚠️ Ethical Considerations

1. **Privacy**: Video/audio data should be anonymized and stored securely
2. **Consent**: Subjects must provide informed consent for analysis
3. **Bias**: Models may have demographic biases - regular auditing required
4. **Interpretation**: Results are probabilistic indicators, not definitive truth
5. **Context**: Consider cultural and situational factors in interpretation

## 🀝 Contributing

1. Fork the repository
2. Create a feature branch (`git checkout -b feature/amazing-feature`)
3. Commit changes (`git commit -m 'Add amazing feature'`)
4. Push to branch (`git push origin feature/amazing-feature`)
5. Open a Pull Request

## πŸ“„ License

MIT License - see LICENSE file for details

## πŸ“§ Contact

For questions or support, please open an issue on GitHub.

---