Classification of TV News Shots with Mining Frequent Frame Composition in News Video Archives

• Project/Area Number: 24650043
• Research Category: Grant-in-Aid for Challenging Exploratory Research
• Allocation Type: Single-year Grants
• Research Field: Media informatics/Database
• Research Institution: National Institute of Informatics
• Principal Investigator: KATAYAMA Norio 国立情報学研究所, コンテンツ科学研究系, 准教授 (60280559)
• Project Period (FY): 2012-04-01 – 2014-03-31
• Project Status: Completed (Fiscal Year 2013)
• Budget Amount: ¥3,900,000 (Direct Cost: ¥3,000,000、Indirect Cost: ¥900,000); Fiscal Year 2013: ¥1,300,000 (Direct Cost: ¥1,000,000、Indirect Cost: ¥300,000); Fiscal Year 2012: ¥2,600,000 (Direct Cost: ¥2,000,000、Indirect Cost: ¥600,000)
• Keywords: マルチメディア / 放送映像アーカイブ

Research Abstract

The aim of our research is to classify TV news shots that are extracted by the cross TV-channel filtering. The cross TV-channel filtering is our formerly proposed method which extracts such news shots that are commonly broadcasted by multiple broadcasters. The classification of TV news shots is based on the video frame composition. With mining frequent frame composition in news video archives both in bottom-up and top-down manners, visual features for the classification have been investigated.