This project provides a flask application for image retrieval. It consists of 3 modules:
Deep Image retrieval, which allows users to perform visual searches both on the query images and any new image.
Spatial Reranking for reranking the results based on local feature matching and finding query regions on the target image.
Locality-sensitive hashing (LSHash) for coping with the large-scale dataset.
Make sure you have Nodejs, Npm and Yarn installed
Download the code and install package
// Download sourcecode
git clone https://github.com/vinhqngo5/CS336-ImageRetrieval.git
// Install required package with Yarn
yarn install
To start frontend run
yarn start
Use conda to create enviroment from environment.yml file
conda env create -f environment.yml
conda activate myenv
Download the code
git clone https://github.com/manhkhanhad/ImageRetrival.git
Setup Image Retrieval module, please follow these instructions
Setup Image Spatial Re-ranking module, please follow these instructions
Start backend
sh start_backend.sh
Go to http://localhost:3000 in your browser and experience.
Manh-Khanh Ngo Huu manhkhanhad
Huy Nguyen akaRainbowShine
Vinh Quang Ngo vinhqngo5
Special thanks to Jerome Revaud, Rafael de Rezende, Cesar de Souza, Diane Larlus, and Jon Almazan at Naver Labs Europe. In this library, we have used the code and pretrained model for Retrieval module from their awesome Image Retrieval repository.
Special thanks to Dmytro Mishkin, Filip Radenovic, Jiri Matas. We have used their code (Image Matching) to make the Spatial Reranking module.
Special thanks to Loreto Parisi for his awesome LSHash library
Give a ⭐️ if this project helped you!
This README was generated with ❤️ by readme-md-generator