Video Processing Pipeline

A lightweight web service for running automated video segmentation and metadata extraction jobs without manual command-line intervention.

How It Works

  • File Ingestion: Users upload video files through a browser interface with progress feedback.
  • Frame Processing: Python workers segment video streams, sample keyframes, and run classification routines on frame sequences using OpenCV.
  • Metadata Output: Generates structured JSON summaries of detected segments, timestamps, and extracted frame properties.

Stack

  • Backend: Python and Flask for HTTP endpoints and dispatching processing scripts.
  • Client: HTML5, CSS, and asynchronous JavaScript for uploads and status polling.
  • Processing: OpenCV and ffmpeg bindings for video file manipulation.