🕵️ Forensic Synopsis Generator & Master Controller

An advanced AI-powered forensic video analysis platform that coordinates multiple state-of-the-art computer vision models running on Kaggle GPUs, processes bounding boxes, computes saliency intervals, and compiles key suspect video tracks.

🚀 Key Features

• Centralized Model Control: Seamlessly orchestrates and triggers three distinct AI models (YOLOv8 + EffNet + Indexing, DeepFace, and Best Retrieval models) on Kaggle Cloud GPUs.
• Smart Suspect Zoom & Track Crops: Extracts suspect tracks and aligns candidate matching crops into clean, uniform galleries using custom blurred-background paddings.
• Temporal Synopsis Generation: Automatically detects target visibility intervals and constructs condensed video summaries (synopses).
• Audio Synchronization: Re-integrates original scene audio with synopses by dynamically cutting and stitching audio streams matching target interval timestamps using FFmpeg.
• Premium Frontend UX: Responsive dark/light UI built with Streamlit, styled with custom CSS card containers, hover micro-animations, and fluid transitions.

---

🛠️ Tech Stack & Requirements

• Frontend UI: Streamlit
• Backend Processing: Python 3.9+, OpenCV, Pandas, SciPy, MoviePy, Kaggle CLI
• Audio/Video Transcoding: FFmpeg
• Cloud GPU Inference: Kaggle API (Kernels Push)

---

⚙️ Setup & Deployment

Local Setup

1. Clone the repository:

   git clone https://github.com/CreatorguyS/Forensic-synopsis-generator.git
   cd Forensic-synopsis-generator

2. Install Python dependencies:

   pip install -r requirements.txt

3. Install system-level dependency ffmpeg.
4. Place your Kaggle API token (kaggle.json) in ~/.kaggle/.
5. Run the web application:

   streamlit run app.py

☁️ Streamlit Community Cloud Deployment

1. Log in to Streamlit Share using GitHub.
2. Select your repository, set the branch to main, and the Entry point to app.py.
3. Open Advanced Settings -> Secrets and configure your Kaggle credentials:

   KAGGLE_USERNAME = "your_kaggle_username"
   KAGGLE_KEY = "your_kaggle_api_key"

4. Click Deploy. Streamlit will automatically read packages.txt to install ffmpeg and requirements.txt for Python packages.