Skip to content

An image retrieval system that utilizes deep learning ResNet for feature extraction, Local Optimized Product Quantization techniques for storage and retrieval, and efficient deployment using Nvidia technologies like TensorRT and Triton Server, all accessible through a FastAPI-powered web API.

License

Notifications You must be signed in to change notification settings

TunggTungg/image_retrieval

Repository files navigation

Image Retrieval

Demo video


Project deployment.

Description

This project aims to build an image retrieval system that utilizes deep learning ResNet for feature extraction, Local Optimized Product Quantization techniques for storage and retrieval, and efficient deployment using Nvidia technologies like TensorRT and Triton Server, all accessible through a FastAPI-powered web API.

Run Locally

Clone the project

  git clone https://github.com/TunggTungg/image_retrieval.git

Go to the project directory

  cd image_retrieval

Run with Docker

  docker-compose up --build

Contributing

Pull requests are welcome. For major changes, please open an issue first to discuss what you would like to change.

Please make sure to update tests as appropriate.

License

MIT

About

An image retrieval system that utilizes deep learning ResNet for feature extraction, Local Optimized Product Quantization techniques for storage and retrieval, and efficient deployment using Nvidia technologies like TensorRT and Triton Server, all accessible through a FastAPI-powered web API.

Topics

Resources

License

Stars

Watchers

Forks