Based on this project, the paper "Design and Implementation of a Real Time Visual Search System on JD E-commerce Platform" has be accepted and publised by International ACM Middleware Conference. https://arxiv.org/abs/1908.07389

The Key features of vearch

  • Large scale vector storage and retrieval

    Vearch is a distributed vector storage and retrieval system which can be easily extend to billions scale.

  • High-performance real time index

    Vearch implements a high-performance,lockless real time vector indexing subsystem that utilizes various optimization techniques to support millisecond vector update and retrieval.

  • Out of box

    End to end one-click deployment. Through the module of plugin, a complete default visual search system can be deployed just with one click. Otherwise, you can easily customize your own image, video or text feature extraction algorithm plugin.

  • Scalar & numeric filter

    Not only vector, vearch can also store scalar data, including numeric and string types. Besides storage, vearch can index numeric types, supporting equal or range filter, to reduce unnecessary computation.

This GIF provides a clear demonstration of the project vearch usage and its internal structure. The use of vearch is mainly divided into three steps. Firstly, create DB and Space, then import your data, and finally you can search on your own dataset.