All projects
June 2026

Best Innovation/Research Project, UTS AI Showcase 2026

PitchSense

An end-to-end football analytics system combining player and ball detection, pitch registration, tracking, and a Streamlit analysis interface.

Overview

Computer vision for football tactics, turning ordinary match footage into pitch-level telemetry without GPS trackers or specialist hardware.

Highlights

  • Won Best Innovation/Research Project at the 2026 UTS AI Showcase.
  • Mapped player and ball movement onto a 105 m x 68 m tactical pitch using YOLO-Pose, ByteTrack, DLT homography, and smoothing.
  • Explored possession, heatmaps, formations, territory, passing networks, pressing, and synchronized video review.

Tools and methods

YOLOYOLO-PoseByteTrackOpenCVStreamlitComputer Vision