TRECVID 2008 - goals, tasks, data, evaluation mechanisms and metrics
Over, Paul and Awad, George M. and Rose, Travis and Fiscus, Jon and Kraaij, Wessel and Smeaton, Alan F. (2008) TRECVID 2008 - goals, tasks, data, evaluation mechanisms and metrics. In: TRECVid 2008, 17-18 November 2008, Gaithersburg, Maryland, USA.
Full text available as:
The TREC Video Retrieval Evaluation (TRECVID) 2008 is a TREC-style video analysis and retrieval evaluation, the goal of which remains to promote progress in content-based exploitation of digital video via open, metrics-based evaluation. Over the last 7 years this effort has yielded a
better understanding of how systems can effectively accomplish such processing and how one can reliably benchmark their performance. In 2008, 77 teams (see Table 1) from various research organizations --- 24 from
Asia, 39 from Europe, 13 from North America, and 1 from Australia --- participated in one or more of five tasks: high-level feature extraction, search (fully automatic, manually assisted, or interactive), pre-production video (rushes) summarization, copy detection, or surveillance event detection. The copy detection and surveillance event detection tasks are being run for the first time in TRECVID.
This paper presents an overview of TRECVid in 2008.
Archive Staff Only: edit this record