Content-based Video Retrieval Using image queries
Kinjal A. Patel, Hiren Patel · International journal of advance research and innovative ideas in education · 2017
Given a query image, the objective is to extract, from a database of videos, the relevant videos to the query image. Here relevant means “visual similarity”. Majority of the queries on Google search want video answers from YouTube or similar sites. Learners prefer information to be presented to them in audio-visual media than text alone. But there is a huge “semantic gap” between what people want, how they can express, and what is available out there. With lack of satisfaction from textual based video retrieval, the idea of content based video retrieval has been the attention for researchers since long time. Our system has integrated different algorithms to propose a new approach for achieving higher efficiency in the field of content based video retrieval.