Point of Interest Recognition and Tracking in Aerial Video during Live Cycling Broadcasts

被引:0
|
作者
Vanhaeverbeke, Jelle [1 ]
Decorte, Robbe [1 ]
Slembrouck, Maarten [1 ]
Van Hoecke, Sofie [1 ]
Verstockt, Steven [1 ]
机构
[1] Ghent Univ imec, IDLab, B-9052 Ghent, Belgium
来源
APPLIED SCIENCES-BASEL | 2024年 / 14卷 / 20期
关键词
landmark recognition; object tracking; computer vision; live processing; aerial video; END;
D O I
10.3390/app14209246
中图分类号
O6 [化学];
学科分类号
0703 ;
摘要
Road cycling races, such as the Tour de France, captivate millions of viewers globally, combining competitive sportsmanship with the promotion of regional landmarks. Traditionally, points of interest (POIs) are highlighted during broadcasts using manually created static overlays, a process that is both outdated and labor-intensive. This paper presents a novel, fully automated methodology for detecting and tracking POIs in live helicopter video streams, aiming to streamline the visualization workflow and enhance viewer engagement. Our approach integrates a saliency and Segment Anything-based technique to propose potential POI regions, which are then recognized using a keypoint matching method that requires only a few reference images. This system supports both automatic and semi-automatic operations, allowing video editors to intervene when necessary, thereby balancing automation with manual control. The proposed pipeline demonstrated high effectiveness, achieving over 75% precision and recall in POI detection, and offers two tracking solutions: a traditional MedianFlow tracker and an advanced SAM 2 tracker. While the former provides speed and simplicity, the latter delivers superior segmentation tracking, albeit with higher computational demands. Our findings suggest that this methodology significantly reduces manual workload and opens new possibilities for interactive visualizations, enhancing the live viewing experience of cycling races.
引用
收藏
页数:20
相关论文
共 7 条
  • [1] A method of robust pedestrian tracking in video sequences based on interest point description
    An, Ming-Shou
    Kang, Dae-Seong
    International Journal of Multimedia and Ubiquitous Engineering, 2015, 10 (10): : 35 - 46
  • [2] Object recognition analysis in mice using nose-point digital video tracking
    Benice, T. S.
    Raber, Jacob
    JOURNAL OF NEUROSCIENCE METHODS, 2008, 168 (02) : 422 - 430
  • [3] AUGMENTING LIVE AERIAL VIDEO IMAGES WITH GIS INFORMATION TO ENHANCE DECISION MAKING PROCESS DURING EMERGENCIES
    Laka-Inurrategi, Maider
    Garcia, Julen
    Alonso, Kevin
    Quartulli, Marco
    4TH INTERNATIONAL CONFERENCE ON CARTOGRAPHY AND GIS, VOL. 1, 2012, : 95 - +
  • [4] DOMINANT SPATIO-TEMPORAL MODULATIONS AND ENERGY TRACKING IN VIDEOS: APPLICATION TO INTEREST POINT DETECTION FOR ACTION RECOGNITION
    Georgakis, Christos
    Maragos, Petros
    Evangelopoulos, Georgios
    Dimitriadis, Dimitrios
    2012 IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP 2012), 2012, : 741 - 744
  • [5] Feasibility of Live Video Feed Transmission from Unmanned Aerial Vehicles for Medical Surveillance During the 2022 Montreal Marathon
    Lafortune, Raphael
    Afram, Eddy
    Grossman, Arielle
    Drolet, Ann-Rebecca
    de Champlain, Francois
    Iannuzzi, David
    Homier, Valerie
    PREHOSPITAL AND DISASTER MEDICINE, 2023, 38 (05) : 617 - 621
  • [6] Keyframe-based recognition and localization during video-rate parallel tracking and mapping
    Castle, R. O.
    Murray, D. W.
    IMAGE AND VISION COMPUTING, 2011, 29 (08) : 524 - 532
  • [7] Toward emotional recognition during HCI using marker-based automated video tracking
    Soderstrom, Ulrik
    Li, Songyu
    Claxton, Harry L.
    Holmes, Daisy C.
    Ranji, Thomas T.
    Santos, Carlos P.
    Westling, Carina E. I.
    Witchel, Harry J.
    PROCEEDINGS OF THE 31ST EUROPEAN CONFERENCE ON COGNITIVE ERGONOMICS: DESIGN FOR COGNITION (ECCE 2019), 2019, : 49 - 52