Reverse video search is similar like reverse image search. In simple words, it takes a video as input to search for similar videos. As we know that video-related tasks are harder to tackle, video models normally do not achieve as high scores as other types of models. However, there are increasing demands in AI applications in video. Reverse video search can effectively discover related videos and improve other applications.
This reverse video search example mainly consists of three notebooks, and I think everyone can learn the basic operations of Towhee and Milvus through the getting started notebook. And the deep dive notebook will make the engine more feasible in production.
In this notebook you will get prerequisites, build and use a basic reverse video search system, visualize sample results, make simple optimizations, and measure the system with performance metrics.
In this notebook, you will learn how to reduce resource usage, speed up system, and ensure stability.