Content-based video indexing for sports applications using integrated multi-modal approach

摘要: To sustain an ongoing rapid growth of video information, there is emerging demand for a sophisticated content-based indexing system. However, current solutions are still immature and lack any standard. This doctoral consists research work based on integrated multi-modal approach sports retrieval. By combining specific features extractable from multiple audio-visual modalities, generic structure events can be detected classified. During browsing retrieval, users will benefit the integration high-level semantic some descriptive mid-level such as whistle close-up view player(s).