Emperor Penguin optimized event recognition and summarization for cricket highlight generation
- PDF / 3,341,000 Bytes
- 15 Pages / 595.276 x 790.866 pts Page_size
- 104 Downloads / 173 Views
LETTER TO THE EDITOR
Emperor Penguin optimized event recognition and summarization for cricket highlight generation Hansa Shingrakhia1,2 · Hetal Patel1,3 Received: 20 March 2020 / Accepted: 10 August 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract Cricket highlight generation is the process of summarizing a full-length video to a shortened form which should preserve the important moments present in the original video. In this paper, a new approach has been proposed for recognizing the key events and summarization. Audio features are initially used for extracting the excitement clips. Then, the important events like replay, players, umpires, spectators, and players gathering are extracted from each clip. Here, a hybrid deep neural network with Emperor Penguin optimization (HDNN-EPO) is proposed for labeling the excitement concepts presented in the cricket video based on the observed events automatically. These labeled concepts are then selected based on the importance degree and concatenated in temporal order to form highlights. The efficiency of the proposed method is proved through the experimental results and it outperforms the other existing approaches. Also, the extracted highlights have been compared with the manually-generated highlights by the sports television channel. Keywords Highlight generation · Excitement · Event-driven · Deep neural network · Emperor Penguin optimization
1 Introduction The automatic analysis of sports events is a major demanding task discussed over the past few years [1]. Detection of events that are reviewed as important for processing as well as offering the chance to realize the reactive measures is termed as event recognition [2]. Sports event recognition has unbelievable potential in the fields of football, cricket, tennis, and so on [3, 4]. The recognition of events keeps the sports fan updated with the up-to-the-minute news as there are several sports matches released every year. Event recognition is technologically advanced to enrich the facility of broadcast thereby giving more data for exact umpiring decisions as well as game analysis [5]. The intensity at which Communicated by T. Yao. * Hansa Shingrakhia [email protected] 1
Gujarat Technological University, Ahmedabad, Gujarat, India
2
ECE Department, Indus University, Ahmedabad, Gujarat, India
3
ECE Department, A.D. Patel Institute of Technology and Engineering, Anand, Gujarat, India
the interest in sports has been increasing extremely leads to nearby contest [6]. Cricket highlights mainly focus on event-driven as well as excitement-based features [7]. The event-driven features include boundaries, sixes, wickets as well as milestones [8]. The excitement-based features include action from crowds, commentators, and players [9]. The key-event is generated from each match as well as the video is shortened according to the user’s convenience [10]. The significant source events are shown by the highlights without consuming too much time and also they keep the sports f
Data Loading...
