How anshuuuuuuuuuuuuu/deepfake-detector-and-ip-tracker works
Project type and purpose
• From the flat file tree (app.py, unmasked.py, an HTML template, CSS files) and README description, this is a small full-stack Python web application whose goal is to allow end-users to upload or stream video, run a “DeepSeek-R1” deepfake-detection model, and surface both detection results and suspicious-account tracking information via external APIs.
File structure
• Top-level files only:
• app.py – the main web server (likely a Flask or similar microframework entry point)
• unmasked.py – the module that wraps or implements the DeepSeek-R1 inference logic
• news_slider.html, style.css, style2.css – frontend assets (HTML + styling)
• startapp.bat – a Windows helper to launch the app
• No database directories or config folders, implying a monolithic service: server + model + API calls all in one codebase.
README for additional insights
• Confirms DeepSeek-R1 model integration
• Mentions “extra tracking apis for suspicious account” – external HTTP services the backend calls
• No built-in data persistence indicated (no DB), likely ephemeral or in-memory