상세정보
학술대회
Pornographic Video Classification Using Fast Motion Features
- 발행일
- 201501
- 출처
- Korea-Japan Joint Workshop on Frontiers of Computer Vision (FCV) 2015, pp.1-4
- 협약과제
- 14MS2700, 스마트 단말용 스트리밍 유해 컨텐츠 차단 기술 개발, 한승완
- 초록
- In this paper, a novel method to automatically classify an pornographic video using fast motion features is proposed. In proposed method, predefined number of clips without a shot change are extracted from an input video and new, fast motion features are computed for each clip. Each clip is given a pornographic possibility based on motion distribution information. If the possibility is bigger than a threshold value, the clip is regarded as an pornographic clip. Finally, the ratio of pornographic clips is computed and the input video is classified as an pornographic video if the ratio is bigger than a threshold ratio.
- KSP 제안 키워드
- Motion distribution, Shot Change, Threshold Value, Video classification, distribution information, fast motion, motion features, novel method