Case study/AI-driven table tennis performance analytics
AI-powered system delivering real-time, frame-level table tennis performance insights
MoreYeahs built a solution that replaces manual, limited video review with fully automated, frame-level gameplay insights from match footage, delivering objective, data-driven metrics that support smarter coaching decisions and clearer gameplay understanding.
- Industry
- Sports Technology & AI Analytics
- Engagement
- Computer Vision-Based Performance Analytics Platform
- Environment
- Professional & Training-Level Table Tennis Match Footage
- Users
- Coaches, Players, Analysts & Broadcasters
Results at a glance
- 01Accurate Player & Table Detection
The system achieved accurate detection of players across all frames and maintained stable identification of table boundaries.
- 02Reliable Ball Tracking
Ball tracking and trajectory mapping were highly reliable, with bounce points successfully extracted for analysis.
- 03Consistent Pose Capture
Player keypoints were consistently detected to capture movement patterns.
01The client
Customer
The client is a sports technology-focused organization exploring AI-driven performance analytics for table tennis at both professional and training levels.
They aim to leverage computer vision and deep learning to extract high-speed gameplay insights from video footage, transforming raw match videos into structured performance data for coaches, players, analysts, and broadcasters.
With a strong interest in innovation, the client sought scalable, real-time analytics to enhance training effectiveness and viewer engagement.
02Challenge
The client faced several obstacles in bringing objective, data-driven analysis to table tennis gameplay:
- 01Limited Frame-Level Analytics
The client lacked granular, frame-level analytics needed to accurately evaluate table tennis gameplay.
- 02Difficult Ball Tracking
The extremely fast-moving ball proved difficult to track with conventional video analysis tools.
- 03Lack of Objective Metrics
Players and coaches had limited access to objective, data-driven performance metrics.
- 04Manual, Inconsistent Analysis
Match analysis relied heavily on manual tagging and annotation, making the process time-consuming and inconsistent.
03Solution
Solution
MoreYeahs built a solution that replaces manual, limited video review with fully automated, frame-level gameplay insights from match footage, delivering objective, data-driven metrics that support smarter coaching decisions and clearer gameplay understanding.
Fully automates frame-level gameplay insights from match footage in place of manual, limited video analysis.
Delivers objective, data-driven metrics that support smarter coaching decisions and clearer gameplay understanding.
Robust ball tracking through combined object detection and segmentation.
Precise bounce point extraction from tracked ball trajectories.
Integrated player pose estimation for biomechanical and movement insight.
04How we delivered
The approach focused on automating table tennis match analysis using advanced computer vision and AI techniques.
- Step 1Ingest
Full HD match videos were ingested into the pipeline for processing.
- Step 2Extract Frames
Frames were extracted from the footage for detailed, frame-level processing.
- Step 3Detect
Object detection identified players, the table, and the ball in each frame.
- Step 4Track
SAM2 segmentation and trajectory analysis tracked ball movement with precision.
- Step 5Estimate
Player keypoints and poses were estimated to capture biomechanical insights and movement patterns.
05Technology
Technology
- Development
- Python
- Object Detection & Segmentation
- YOLOv8SAM2
- Pose Estimation
- MediaPipe
- Frame Processing & Visualization
- OpenCV
- Data Structuring & Export
- NumPyPandas
06Results
The automated pipeline delivered reliable, high-quality analytics across detection, tracking, and reporting.
- Accurate Player & Table Detection
The system achieved accurate detection of players across all frames and maintained stable identification of table boundaries.
- Reliable Ball Tracking
Ball tracking and trajectory mapping were highly reliable, with bounce points successfully extracted for analysis.
- Consistent Pose Capture
Player keypoints were consistently detected to capture movement patterns.
- Structured, Exportable Data
All data was structured and exported for downstream analytics and performance evaluation.
- Actionable Visual Insights
High-quality annotated visuals provided clear, actionable insights from the match footage.
07Business impact
Business Impact
Beyond a single proof of concept, the work shows how computer vision and deep learning can generate lasting value from match footage, converting raw video into actionable data for coaches, analysts, and broadcasters.
Coaches, analysts, and broadcasters gain objective, data-driven performance metrics in place of manual review.
Automated player and ball detection, trajectory mapping, and pose estimation convert raw match video into actionable data.
The platform lays a strong technical foundation for future sports analytics and intelligent coaching applications.
The approach highlights the potential for AI-enhanced broadcasting and real-time game analysis.
More projects like this
All case studiesFacing a similar challenge?
Tell us what you need. A technical lead will reply within one business day with how we would approach it.
- Reply within one business day
- Free 30-minute call
- NDA before you share data