<title>Automated analysis and annotation of basketball video</title>

Abstract

Automated analysis and annotation of video sequences are important for digital video libraries, content-based video browsing and data mining projects. A successful video annotation system should provide users with useful video content summary in a reasonable processing time. Given the wide variety of video genres available today, automatically extracting meaningful video content for annotation still remains hard by using current available techniques. However, a wide range video has inherent structure such that some prior knowledge about the video content can be exploited to improve our understanding of the high-level video semantic content. In this paper, we develop tools and techniques for analyzing structured video by using the low-level information available directly from MPEG compressed video. Being able to work directly in the video compressed domain can greatly reduce the processing time and enhance storage efficiency. As a testbed, we have developed a basketball annotation system which combines the low-level information extracted from MPEG stream with the prior knowledge of basketball video structure to provide high level content analysis, annotation and browsing for events such as wide-angle and close-up views, fast breaks, steals, potential shots, number of possessions and possession times. We expect our approach can also be extended to structured video in other domains.Keywords: structured video analysis, basketball video content annotation, browsing, indexing, MPEG compressed data, wide-angle and close-up camera views, intra-coded block count.

<title>Automatic acquisition of motion trajectories: tracking hockey players</title>

“…Although there have been video annotation systems in the domain of soccer [3,7,17,38], basketball [29] and football [11], to the best of our knowledge, there have not been any automatic video annotation systems in hockey.…”

“…For instance, Saur et. al [29] develop a system for automated analysis and annotation of basketball video. Their system provides fast and efficient analysis of basketball video by using only informa- With an accurate registration of the globally consistent football sequence, they conduct "closed-world" analysis to track football players in complex scenes where "closed-world" is defined as "a region of space and time in which the specific context is adequate to determine all possible objects present in that region."…”

Automatically selecting shots for action movie trailers

“…For example, Assfalg et al [2] detect free kicks, penalties and corner kicks in soccer matches primarily using camera motion and a hidden Markov model, while [7] look for high motion portions of soccer videos as these typically indicate an important event. [28,24], analyse the motion information in basketball video in order to determine which team is currently in the attacking part of the court, and use this to detect scores. There have been similar approaches in other sports such as tennis [15], American football [19], motor sports [26], cricket [16] and baseball [9].…”

Event detection in field sports video using audio-visual features and a support vector Machine

“…Nepal et al [8] describe audio-visual techniques for the detection of basketball goals, while Zhang and Ellis [9] present an audio only approach. Further instances of automated basketball video indexing can be found in [10] and [11]. Audio and/or video based analyses for event detection toward summarization or indexing, can be found in the literature for a variety of sports genres; formula-1 [12]; baseball [13]- [15]; cricket [16]; tennis [17]- [19]; American football [20], and Gaelic football [21].…”