Image analysis and interpretation for semantics categorization in baseball video

The semantic information in videos is useful for content-based video retrieval and summarization. Traditional image/video understanding is formulated in terms of low-level features describing the structure and intensity of the input image/video. The way to generate the high-level knowledge such as common sense and human perceptual knowledge is one of the most difficult problems This paper attempts to bridge this gap through the integration of image analysis algorithms and multi-level semantic network (SN) to interpret the semantic meaning of the baseball video.