Open Access Open Access  Restricted Access Subscription or Fee Access

A Review of Video Shot Boundary Detection (SBD) technique

Shailender Gupta, Shekhar Dhiman, Alisha Makkar, Divya Khanna, Ashima Arora

Abstract


The first step in effective video management system which provides the facility of content based video search, analysis and retrieval system is Video Shot Boundary Detection (SBD). A shot is a sequence of frames obtained from a single camera without interrupting it. A video contains a lot of shots that are accompanied with transitions from one shot to another. When shot transition occurs over a few number of frames, it is called Cut transition and if transition occurs over a relatively larger number of frames, it is called Gradual transition. The inspiration for the SBD scheme is from real time based problems. Several SBD techniques have been proposed over the years which yield satisfactory results for Cut transition but Gradual transition detection still have scope for improvement. In this paper, some of the recent benchmark SBD techniques are implemented and compared on the basis of Performance Metrics such as Precision, Recall, F1 score and Time Complexity.


Keywords


Adaptive Threshold; Pattern Matching; Singular Value Decomposition (SVD); Speeded Up Robust Feature (SURF); Walsh Hadamard Transform (WHT)

Full Text:

PDF

Refbacks

  • There are currently no refbacks.