Real-Time Tennis Computer Vision Project at 30+ FPS

Melvin Vivas · X video post · 2026-03-26 · Open on X

Topics: Portfolio Projects, Programming & ML Foundations · Level: intermediate

Summary

A computer vision portfolio project for tennis that detects the player, ball and racquet and estimates player poses. It uses exclusion zones to remove false positives and interpolation plus smoothing to draw clean ball trails, all at over 30 FPS.

Key points

• Detects player, ball and racquet.
• Estimates player poses.
• Exclusion zones remove false-positive detections.
• Interpolation and smoothing produce clean ball trails.
• Runs in real time at over 30 FPS.

Try this

• Build a sports-analytics CV pipeline: object detection, pose estimation, exclusion zones and smoothed ball tracking running in real time.