Abrupt shot boundary detection based on averaged two-dependence estimators learning
Conference proceedings article
ผู้เขียน/บรรณาธิการ
กลุ่มสาขาการวิจัยเชิงกลยุทธ์
ไม่พบข้อมูลที่เกี่ยวข้อง
รายละเอียดสำหรับงานพิมพ์
รายชื่อผู้แต่ง: Tippaya S., Sitjongsataporn S., Tan T., Chamnongthai K.
ผู้เผยแพร่: Hindawi
ปีที่เผยแพร่ (ค.ศ.): 2015
หน้าแรก: 522
หน้าสุดท้าย: 526
จำนวนหน้า: 5
ISBN: 9781479944163
นอก: 0146-9428
eISSN: 1745-4557
ภาษา: English-Great Britain (EN-GB)
บทคัดย่อ
Video shot boundary detection is the process of automatically detecting the meaningful boundary in video data. It becomes an essential pre-processing step to video analysis, summarisation and other content-based retrieval. Video frame feature representation also plays an important role in the process where it directly affects to the performance of the system. Histogram dissimilarity-based with the pre-processed features scheme are proposed to represent the temporal characteristic in videos. Motivated by the practical applications with moderate computational time, supervised abrupt shot boundary detection with averaged two-dependence estimators probabilistic classification learning scheme is proposed in this paper. The performance evaluation is performed by TRECVID 2007 videos dataset containing various types of video category. The performance of the proposed scheme can be expressed in terms of precision and recall to detect the correct abrupt video shot. ฉ 2014 IEEE.
คำสำคัญ
Abrupt shot boundary detection, averaged twodependence estimators, data dimensionality reduction, independent component analysis, probabilistic learning