Loading article...
Loading article...
Stream/Area:
-
Subject:
-
Paper ID:
IJSREM67649
Volume:
10 |
10 |
Oct 2026
Pages:
1-3
Publication Frequency:
Monthly (12 issues per year)
License:International Journal of Scientific Research in Engineering and Management follows the Creative Commons Attribution 4.0 (CC BY 4.0) License, allowing free use and distribution with proper attribution.
Computer Science and Engineering
Don Bosco Institute of Technology, Karnataka, India
Computer Science and Engineering
Don Bosco Institute of Technology, Karnataka, India
Computer Science and Engineering
Don Bosco Institute of Technology, Karnataka, India
Computer Science and Engineering
Don Bosco Institute of Technology, Karnataka, India
Computer Science and Engineering
Don Bosco Institute of Technology, Karnataka, India
High-speed projectile tracking and automated officiating in baseball pose significant computer vision challenges because of severe motion blur, low object pixel resolution, and dynamic background occlusions. Traditional single-frame object detection models routinely fail under these conditions owing to an absence of temporal continuity. This paper presents AcuStrike, an automated pitch analysis and adjudication framework powered by the TrackNet V5 deep spatio-temporal architecture. AcuStrike formulates ball localization as consecutive multi-frame probability heatmap regression using a hybrid deep convolutional neural network (CNN) integrated with Recurrent Spatio-Temporal Residual (RSTR) Transformer attention. Sub-pixel spatial coordinates decoded from peak activations are fitted with continuous cubic splines and coupled with a physics-grounded Statcast kinematic engine that calculates release-to-plate velocities. Crossing coordinates at home plate are mapped to an automated 3x3 K-Zone grid classifier, which adjudicates pitches into discrete zones or directional misses. Empirical evaluations on real-world 60 and 120 frames-per-second (FPS) high-speed behind-the-pitcher video captures demonstrate reliable trajectory reconstruction, accurate velocity computation up to 124.5 km/h, and objective real-time officiating calls.
KEYWORDS
The AcuStrike framework demonstrates that TrackNet V5 spatio-temporal heatmap estimation effectively overcomes severe motion streak blur and background interference in high-velocity baseball tracking without requiring expensive multi-camera stadium rigs. By integrating multi-frame temporal reasoning, cubic spline trajectory fitting, Statcast kinematics, and an automated 3x3 K-Zone classifier, AcuStrike delivers low-cost, objective pitch adjudication and broadcast-grade visual analytics from monocular video feeds.
1. Fukuta, H., Nakayama, K., Watanabe, H.: Three-Dimensional Baseball Strike Judgement by Monocular Video from Umpire’s Viewpoint. IEEE GCCE (2023). 2. Sugizaki, Y., Nakazato, J., Tsukada, M., Esaki, H.: Umpire Assistance System in Baseball Game. IoT 2023 (2023). 3. Major League Baseball: Strike Zone - Glossary - MLB.com. Official MLB Rules (2023). 4. Lee, K., Han, K., Ko, J.: Analyzing the Impact of the Automatic Ball Strike System in Professional Baseball. arXiv:2407.15779 (2024). 5. Hsieh, J.: Neural Network-Based Tracking and 3D Reconstruction of Baseball Pitch Trajectories from Single-View 2D Video. J. Sports Eng. Technol. (2024). 6. Akatsu, M.: Synthesis of Knowledge for Realizing Society 5.0. Transdisciplinary Fed. Sci. Tech. 14(1), 3–6 (2020). 7. Aroganam, G., Manivannan, N., Harrison, D.: Review on Wearable Technology Sensors Used in Consumer Sport Applications. Sensors 19(9) (2019). 8. Attaran, M., Celik, B. G.: Digital Twin: Benefits, Use Cases, Challenges, and Opportunities. Decis. Anal. J. 6 (2023). 9. MacMahon, C., Starkes, J. L.: Contextual Influences on Baseball Ball-Strike Decisions in Umpires, Players, and Controls. J. Sports Sci. (2008). 10. Chen, J., Watanabe, R., Nonaka, K., et al.: Fast Free-Viewpoint Video Synthesis Algorithm for Sports Scenes. IEEE (2019).
Publication Date
10/3/2026
Peer Review Type
Double-Blind Peer Review
Review Rounds: 2
Documents
Article Views
0
Total views since publication
PDF Downloads
9
Total downloads since publication
Geographic Reach
5
Digital Preservation: This article is permanently archived on Zenodo and complies with all data preservation standards across major indexing platforms.
Plagiarism Screening
This manuscript has been checked for plagiarism using TurnitinPlagiarismChecker software.
Conflict of Interest
The authors declare no conflict of interest related to this research work.
Funding Information
No funding information available for this research.
Related research in the same subject area
Loading similar articles...