Large-Scale Video Database Retrieval via Visual Recommendation 12+

📓 Motivated by Google's great success on text document retrieval, researchers began to build a new generation of video retrieval systems supporting semantic video retrieval via keywords. But these systems are unable to provide satisfactory results because of several challenging problems. First, existing semantic understanding techniques are still immature, and their performance is not good enough to enable keyword-based retrieval. Second, the mismatch between visual concepts and keywords prevents any keyword-based search engines from retrieving visual concepts that are difficult to represent in language. Third, users may not have a clear idea of their needs at the beginning for many multimedia queries. Therefore, they may not be able to represent their preference with keywords. To resolve these problems, we have proposed a novel systematic solution via visual recommendation. It integrates the latest achievements of semantic video analysis, knowledge discovery and visual analytics and optimized all components toward a single target. As a result, the proposed system is able to implement more efficient and intuitive video retrieval.