A sports video browsing and retrieval system based on multimodal analysis: SportsBR

摘要

An advanced sports video browsing and retrieval system based on multimodal analysis, SportsBR, is proposed in this work. Its main features include event-based sports video browsing and keyword-based sports video retrieval. The paper first defines the basic structure of our SportsBR system, and then introduces a novel approach that integrates multimodal analysis, such as visual streams analysis, speech recognition, speech signal processing and text extraction to realize event-based video clips selection. The experimental results for sports video of world cup football games indicate that multimodal analysis is effective for video browsing and retrieval by quickly browsing event-based video clips and inputting keywords according to a predefined sports vocabulary database. The system is proved to be helpful and effective for the overall understanding of the sports video content.