A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash
Sports pose a unique challenge for high-speed, unobtrusive, uninterrupted motion tracking due to speed of movement and player occlusion, especially in the fast and competitive sport of squash. The objective of this study is to use video tracking techniques to quantify kinematics in elite-level squas...
| Main Authors: | , , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
MDPI AG
2020-12-01
|
| Series: | Applied Sciences |
| Subjects: | |
| Online Access: | https://www.mdpi.com/2076-3417/10/24/8793 |
| _version_ | 1797545302026092544 |
|---|---|
| author | Maria Martine Baclig Noah Ergezinger Qipei Mei Mustafa Gül Samer Adeeb Lindsey Westover |
| author_facet | Maria Martine Baclig Noah Ergezinger Qipei Mei Mustafa Gül Samer Adeeb Lindsey Westover |
| author_sort | Maria Martine Baclig |
| collection | DOAJ |
| description | Sports pose a unique challenge for high-speed, unobtrusive, uninterrupted motion tracking due to speed of movement and player occlusion, especially in the fast and competitive sport of squash. The objective of this study is to use video tracking techniques to quantify kinematics in elite-level squash. With the increasing availability and quality of elite tournament matches filmed for entertainment purposes, a new methodology of multi-player tracking for squash that only requires broadcast video as an input is proposed. This paper introduces and evaluates a markerless motion capture technique using an autonomous deep learning based human pose estimation algorithm and computer vision to detect and identify players. Inverse perspective mapping is utilized to convert pixel coordinates to court coordinates and distance traveled, court position, ‘T’ dominance, and average speeds of elite players in squash is determined. The method was validated using results from a previous study using manual tracking where the proposed method (filtered coordinates) displayed an average absolute percent error to the manual approach of 3.73% in total distance traveled, 3.52% and 1.26% in average speeds <9 m/s with and without speeds <1 m/s, respectively. The method has proven to be the most effective in collecting kinematic data of elite players in squash in a timely manner with no special camera setup and limited manual intervention. |
| first_indexed | 2024-03-10T14:13:31Z |
| format | Article |
| id | doaj.art-42a6fbe759a5462b9430c4c3665ea9ff |
| institution | Directory Open Access Journal |
| issn | 2076-3417 |
| language | English |
| last_indexed | 2024-03-10T14:13:31Z |
| publishDate | 2020-12-01 |
| publisher | MDPI AG |
| record_format | Article |
| series | Applied Sciences |
| spelling | doaj.art-42a6fbe759a5462b9430c4c3665ea9ff2023-11-20T23:57:48ZengMDPI AGApplied Sciences2076-34172020-12-011024879310.3390/app10248793A Deep Learning and Computer Vision Based Multi-Player Tracker for SquashMaria Martine Baclig0Noah Ergezinger1Qipei Mei2Mustafa Gül3Samer Adeeb4Lindsey Westover5Department of Electrical and Computing Engineering, University of Calgary, Calgary, AB T2N 1N4, CanadaDepartment of Computing Science, University of Alberta, Edmonton, AB T6G 2E1, CanadaDepartment of Civil and Environmental Engineering, University of Alberta, Edmonton, AB T6G 2E1, CanadaDepartment of Civil and Environmental Engineering, University of Alberta, Edmonton, AB T6G 2E1, CanadaDepartment of Civil and Environmental Engineering, University of Alberta, Edmonton, AB T6G 2E1, CanadaDepartment of Mechanical Engineering, University of Alberta, Edmonton, AB T6G 2G8, CanadaSports pose a unique challenge for high-speed, unobtrusive, uninterrupted motion tracking due to speed of movement and player occlusion, especially in the fast and competitive sport of squash. The objective of this study is to use video tracking techniques to quantify kinematics in elite-level squash. With the increasing availability and quality of elite tournament matches filmed for entertainment purposes, a new methodology of multi-player tracking for squash that only requires broadcast video as an input is proposed. This paper introduces and evaluates a markerless motion capture technique using an autonomous deep learning based human pose estimation algorithm and computer vision to detect and identify players. Inverse perspective mapping is utilized to convert pixel coordinates to court coordinates and distance traveled, court position, ‘T’ dominance, and average speeds of elite players in squash is determined. The method was validated using results from a previous study using manual tracking where the proposed method (filtered coordinates) displayed an average absolute percent error to the manual approach of 3.73% in total distance traveled, 3.52% and 1.26% in average speeds <9 m/s with and without speeds <1 m/s, respectively. The method has proven to be the most effective in collecting kinematic data of elite players in squash in a timely manner with no special camera setup and limited manual intervention.https://www.mdpi.com/2076-3417/10/24/8793video trackingsports broadcast analysisplayer identificationkinematicsracquet sports |
| spellingShingle | Maria Martine Baclig Noah Ergezinger Qipei Mei Mustafa Gül Samer Adeeb Lindsey Westover A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash Applied Sciences video tracking sports broadcast analysis player identification kinematics racquet sports |
| title | A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash |
| title_full | A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash |
| title_fullStr | A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash |
| title_full_unstemmed | A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash |
| title_short | A Deep Learning and Computer Vision Based Multi-Player Tracker for Squash |
| title_sort | deep learning and computer vision based multi player tracker for squash |
| topic | video tracking sports broadcast analysis player identification kinematics racquet sports |
| url | https://www.mdpi.com/2076-3417/10/24/8793 |
| work_keys_str_mv | AT mariamartinebaclig adeeplearningandcomputervisionbasedmultiplayertrackerforsquash AT noahergezinger adeeplearningandcomputervisionbasedmultiplayertrackerforsquash AT qipeimei adeeplearningandcomputervisionbasedmultiplayertrackerforsquash AT mustafagul adeeplearningandcomputervisionbasedmultiplayertrackerforsquash AT sameradeeb adeeplearningandcomputervisionbasedmultiplayertrackerforsquash AT lindseywestover adeeplearningandcomputervisionbasedmultiplayertrackerforsquash AT mariamartinebaclig deeplearningandcomputervisionbasedmultiplayertrackerforsquash AT noahergezinger deeplearningandcomputervisionbasedmultiplayertrackerforsquash AT qipeimei deeplearningandcomputervisionbasedmultiplayertrackerforsquash AT mustafagul deeplearningandcomputervisionbasedmultiplayertrackerforsquash AT sameradeeb deeplearningandcomputervisionbasedmultiplayertrackerforsquash AT lindseywestover deeplearningandcomputervisionbasedmultiplayertrackerforsquash |