Frame fushion
Description
Utilizing Cloudflare Workers and Pages with TypeScript to build frame-fushion
Features
- Video Upload: Users can upload videos from their local machine for analysis.
- Frame Analysis: The tool analyzes individual frames of the video to extract and synthesize key information.
- Object Detection: Detects and tracks objects present in the video frames.
- Scene Analysis: Analyzes scenes to identify different environments or settings in the video.
- Data Visualization: Provides visualizations of the analysis results for easier interpretation.
Prerequisites
- Node.js: Ensure that Node.js is installed on your machine.
- Cloudflare Account: Sign up for a Cloudflare account if you don't have one already.
Getting Started
- Clone the Repository:
git clone https://github.com/ezecodes/frame-fushion.git and cd frame-fushion
- Install dependencies
npm install
- Wrangler.toml
Append the following bindings to your
wrangler.toml file
- Start the Development Server
Run
npm run dev to start the Vite server
Run npm run preview to start wrangler
wrangler.toml
name =
pages_build_output_dir =
compatibility_date =
compatibility_flags = ["nodejs_compat"]
[ai]
binding = "AI"
[[d1_databases]]
binding = // available in your Worker on env.DB
database_name =
database_id =
