Web UI Usage Guide

This guide covers how to install, run, and use the Video Analyzer Web UI.

Prerequisites

• Python 3.8 or higher.
• The video-analyzer package must be installed and accessible in your environment.
• FFmpeg must be installed on the system where the server is running.

Installation

Navigate to the root directory of the repository and install the UI package using pip:

# From the project root
pip install ./video-analyzer-ui

# Or, for development
pip install -e ./video-analyzer-ui

This will install the UI and its dependencies, including Flask. It also sets up the video-analyzer-ui command.

Running the Server

You can run the server in two modes: development and production.

Development Mode

This mode is ideal for local use and development. It enables auto-reloading when code changes and provides detailed debug output.

video-analyzer-ui --dev

By default, the server will be accessible at http://localhost:5000.

Production Mode

For broader access (e.g., within a local network), you can run the server bound to all network interfaces.

video-analyzer-ui --host 0.0.0.0 --port 5000

Command-Line Options

Option	Description	Default
--dev	Enable development mode with auto-reload.	False
--host	The host address to bind the server to.	localhost
--port	The port number to listen on.	5000
--log-file	Path to a file for logging output.	None

Using the Interface

1. Open in Browser: Navigate to the server's address (e.g., http://localhost:5000).

2. Upload a Video: Drag and drop a video file (.mp4, .avi, .mov, .mkv) onto the upload area, or click to open a file selection dialog.

3. Configure Analysis: Once uploaded, the configuration form will appear.

   • Select your desired client (Ollama or OpenAI API).
   • Fill in the required fields for your chosen client (e.g., API Key for OpenAI).
   • Adjust other parameters like the Whisper model, max frames, or add a custom prompt.
   • The Command Preview box will update in real-time to show the command that will be executed.

4. Start Analysis: Click the "Start Analysis" button.

5. Monitor Progress: The view will switch to the output section. The raw output from the video-analyzer command will be streamed to the page, so you can monitor its progress.

6. Download Results: When the analysis completes successfully, a "Download Results" button will appear. Click it to save the analysis.json file.

7. New Analysis: Click the "New Analysis" button to reset the interface and start over with a new video. This will also clean up the files from the previous session on the server.