Video Captioning API

This project is a FastAPI-based API for generating captions for videos. It utilizes a pre-trained model from Hugging Face's model hub to generate captions for uploaded videos.

Installation

1. Clone the repository:

   git clone https://github.com/codedpro/video-captioning-api.git

2. Navigate to the project directory:

   cd video-captioning-api

3. Install the required dependencies:

   pip install -r requirements.txt

About

This API generate captions for uploaded videos. It follows a simple workflow:

1. Video Upload: Users can upload a video file.
2. Video Processing: The uploaded video is processed to extract frames.
3. Caption Generation: The extracted frames are used as input to a pre-trained model to generate captions.
4. Response: The generated captions are returned as JSON response.

The model used for caption generation is microsoft/git-base-vatex from the Hugging Face model hub.

Usage

To use this API, send a POST request to the /caption_video/ endpoint with the video file as form data. Here's an example using curl:

curl -X 'POST' \
  'http://localhost:8000/caption_video/' \
  -H 'accept: application/json' \
  -H 'Content-Type: multipart/form-data' \
  -F 'file=@/path/to/your/video.mp4'

Replace /path/to/your/video.mp4 with the path to your video file.

API Endpoints

/caption_video/ [POST]

• Description: Generate captions for the uploaded video.
• Request Body: Form data with a single field file containing the video file.
• Response: JSON response containing the generated captions.

Credits

• FastAPI
• PyAV
• Hugging Face Transformers

Dev

This Project Developed by @Coded_Pro