Novel automated detection of sports activities using shadow videos

被引:1
|
作者
Barua, Prabal Datta [1 ,2 ,3 ,4 ,5 ,6 ,7 ,8 ,9 ]
Tuncer, Turker [10 ]
Dogan, Sengul [10 ]
Ooi, Chui Ping [11 ]
Acharya, Rajendra U. [12 ]
机构
[1] Cogninet Australia, Sydney, NSW 2010, Australia
[2] Univ Southern Queensland, Sch Business Informat Syst, Toowoomba, Australia
[3] Univ Technol Sydney, Fac Engn & Informat Technol, Sydney, NSW 2007, Australia
[4] Australian Int Inst Higher Educ, Sydney, NSW 2000, Australia
[5] Univ New England, Sch Sci & Technol, Armidale, Australia
[6] Taylors Univ, Sch Biosci, Subang Jaya, Malaysia
[7] SRM Inst Sci & Technol, Sch Comp, Kattankulathur, India
[8] Kumamoto Univ, Sch Sci & Technol, Kumamoto, Japan
[9] Univ Sydney, Sydney Sch Educ & Social Work, Sydney, Australia
[10] Firat Univ, Technol Fac, Dept Digital Forens Engn, Elazig, Turkiye
[11] Singapore Univ Social Sci, Sch Sci & Technol, Singapore 599494, Singapore
[12] Univ Southern Queensland, Sch Math Phys & Comp, Springfield, Australia
关键词
Sports activities detection using shadow; Video processing; Transfer learning; Chi2 feature selection; LOSO validation; PHYSICAL-ACTIVITY; CLASSIFICATION; EXERCISE; FITNESS; LIFE;
D O I
10.1007/s11042-023-17407-1
中图分类号
TP [自动化技术、计算机技术];
学科分类号
0812 ;
摘要
The aim of this work is to investigate the use of shadow videos for daily sports activity detection and to contribute to the emerging field of shadow-based classification in machine learning. A novel deep feature engineering model is proposed, and a new shadow video dataset is collected to validate the proposed model. Furthermore, the use of shadow videos ensures privacy protection for individuals. To evaluate the proposed system, a dataset comprising five sports activities (i.e., squat, steady run, standing butterfly, overhead side bend, and knee lift) recorded from 33 participants is used. The proposed model works in the following way: (i) videos are divided into frames and aggregated into non-overlapping blocks of six frames to create images, (ii) deep features are extracted from three fully connected layers of the pre-trained AlexNet, resulting in 4096, 4096, and 1000 features from fc6, fc7, and fc8 layers, respectively. These three feature vectors are then merged to generate a final feature vector with a length of 9192 (= 4096 + 4096 + 1000). (iii) The Chi2 selector is employed to select the most distinctive 1000 features in the feature selection phase, and (iv) the support vector machine (SVM) with leave-one-subject-out (LOSO) validation is used to classify the five sports activities. The proposed deep features coupled with Chi2 based model achieved a classification accuracy of 88.49% using the SVM classifier with LOSO cross-validation (CV) on our collected dataset.
引用
收藏
页码:44933 / 44954
页数:22
相关论文
共 50 条
  • [41] A Novel Shadow Detection Method using Fuzzy Rule based Model
    Das, Sonia
    Meher, Sukadev
    PROCEEDINGS OF THE 2017 IEEE 15TH STUDENT CONFERENCE ON RESEARCH AND DEVELOPMENT (SCORED), 2017, : 493 - 498
  • [42] Real-time Detection of Activities in Untrimmed Videos
    Gleason, Joshua
    Castillo, Carlos D.
    Chellappa, Rama
    2020 IEEE WINTER CONFERENCE ON APPLICATIONS OF COMPUTER VISION WORKSHOPS (WACVW), 2020, : 117 - 125
  • [43] ONLINE, SIMULTANEOUS SHOT BOUNDARY DETECTION AND KEY FRAME EXTRACTION FOR SPORTS VIDEOS USING RANK TRACING
    Abd-Almageed, Wael
    2008 15TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING, VOLS 1-5, 2008, : 3200 - 3203
  • [44] Automated Detection of Learning Styles using Online Activities and Model Indicators
    Lestari, Alia
    Lawi, Armin
    Thamrin, Sri Astuti
    Hidayat, Nurul
    INTERNATIONAL JOURNAL OF ADVANCED COMPUTER SCIENCE AND APPLICATIONS, 2024, 15 (06) : 594 - 604
  • [45] SportsHHI: A Dataset for Human-Human Interaction Detection in Sports Videos
    Wu, Tao
    He, Runyu
    Wu, Gangshan
    Wang, Limin
    2024 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2024, : 18537 - 18546
  • [46] Automated detection of electrocautery instrument in videos of open neck procedures using YOLOv3
    Deng, Tingyan
    Gulati, Shubham
    Rodriguez, William
    Dawant, Benoit M.
    Langerman, Alexander
    2021 43RD ANNUAL INTERNATIONAL CONFERENCE OF THE IEEE ENGINEERING IN MEDICINE & BIOLOGY SOCIETY (EMBC), 2021, : 2071 - 2074
  • [47] Real-Time Tracking-by-Detection in Broadcast Sports Videos
    Sverrisson, Sigurdur
    Grancharov, Volodya
    Pobloth, Harald
    IMAGE ANALYSIS, 2019, 11482 : 399 - 411
  • [48] Unsupervised Temporal Feature Aggregation for Event Detection in Unstructured Sports Videos
    Chaudhury, Subhajit
    Kimura, Daiki
    Vinayavekhin, Phongtharin
    Munawar, Asim
    Tachibana, Ryuki
    Ito, Koji
    Inaba, Yuki
    Matsumoto, Minoru
    Kidokoro, Shuji
    Ozaki, Hiroki
    2019 IEEE INTERNATIONAL SYMPOSIUM ON MULTIMEDIA (ISM 2019), 2019, : 9 - 16
  • [49] Automatic Detection of Headlines in Temporally Aggregated TV Sports News Videos
    Choros, Kazimierz
    2013 8TH INTERNATIONAL SYMPOSIUM ON IMAGE AND SIGNAL PROCESSING AND ANALYSIS (ISPA), 2013, : 147 - 152
  • [50] Shadow Detection and Removal Using a Shadow Formation Model
    Shedlovska, Yana I.
    Hnatushenko, Volodymyr V.
    PROCEEDINGS OF THE 2016 IEEE FIRST INTERNATIONAL CONFERENCE ON DATA STREAM MINING & PROCESSING (DSMP), 2016, : 187 - 190