Basketball Flight Trajectory Tracking using Video Signal Filtering

Zhang, Botao; Zhang, Yiheng; Alshawi, Bandar; Alturki, Ryan

doi:10.1007/s11036-023-02253-0

Cited by 3 publications

(2 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For the positive sample, assuming the coordinates of the center point of the true ball labeling box and the size, use Equations ( 10) to (11) to obtain the horizontal and vertical clipping offsets and Equations ( 12) and ( 13) to obtain the corresponding clipping boundaries. A ball patch can be obtained according to Equations (10) to (13). Repeating the above process can obtain a set of multiple positive samples.…”

Section: Heuristic Samplingmentioning

confidence: 99%

“…Zhao et al [12] combined the MeanShift algorithm with adaptive object area size to improve Kalman filter tracking estimation, improving tracking accuracy and speed. Zhang et al [13] utilize Harris corner detection, SURF feature extraction, and a particle swarm optimization algorithm to improve basketball tracking results. Roman et al [14] preset a fixed basketball size and search for basketball position points in the scene using RANSAC (Random Sample Consensus).…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Ball Tracking Based on Multiscale Feature Enhancement and Cooperative Trajectory Matching

Han,

Wang,

Wang

2024

Applied Sciences

View full text Add to dashboard Cite

Most existing object tracking research focuses on pedestrians and autonomous driving while ignoring sports scenes. When general object tracking models are used for ball tracking, there are often problems, such as detection omissions due to small object sizes and trajectory loss due to occlusion. To address these challenges, we propose a ball detection and tracking model called HMMATrack based on multiscale feature enhancement and multilevel collaborative matching to improve ball-tracking results from the entire process of sampling, feature extraction, detection, and tracking. It includes a Heuristic Compound Sampling Strategy to deal with tiny sizes and imbalanced data samples; an MNet-based detection module to improve the ball detection accuracy; and a multilevel cooperative matching and automatic trajectory correction tracking algorithm that can quickly and accurately correct the ball’s trajectory. We also hand-annotated SportsTrack, a ball-tracking dataset containing soccer, basketball, and volleyball scenes. Extensive experiments are conducted on the SportsTrack, demonstrating that our proposed HMMATrack model outperforms other representative state-of-the-art models in ball detection and tracking.

show abstract

Section: Heuristic Samplingmentioning

confidence: 99%