Interactive Robust Multitudinous Visual Search on Mobile Devices?

This paper describes a novel interactive multitudinous visual search system for multiple type of input e.g. text, image, voice or any combination of that inputs using Geotag on mobile devices. The system, jointly search for Image, Speech, And text, takes full benefits of the multiple type of input and get best user interactions on mobile devices. It is designed for users who have number of imaginations in their mind but unable to describe their expectations and cannot get desired results for entered query. The proposed system is able to achieve best search performance and is ten times faster. We are developing a system to search result using Geotag by which tracing location which will show best desired result (location based search), with all possible similar images. This is the main advantage of proposed system. We are adding large vocabulary (trained dataset) for getting maximum interesting result for multitudinous input.