Sports intelligence & tracking

TSX Sports AI

Computer vision pipelines for pitch registration, continuous player tracking, and automated tactical event reconstruction.

Football performance analyst reviewing conceptual player paths from live match footage

Conceptual supporting visual based on the live product workflow; not a product screenshot.

Project evidence

Published project record

Published project evidence from Alector Lab. Alector Lab project record; outcome labels are descriptive rather than independent performance measurements.

Measurement period: Current published product scope

Evidence method: Capabilities are described from the public product experience; outcome labels are not independent performance measurements.

Review the project source
No taggingAutomated workflow

From uploaded footage to coaching insight

SidelineCamera-first design

Built for the fixed-pivot footage many clubs already record

The problem

Many clubs already record matches but still depend on manual tagging or broadcast-oriented tools. That leaves useful player behaviour, awareness, pressure, and decision context buried inside video.

The approach

The pipeline follows players through camera movement, calibrates the pitch into real-world coordinates, reconstructs match events, and computes player and team measures including scan frequency and composite performance indicators.

Implementation

Clubs upload footage, follow analysis progress, and receive match views built for coaching review. The product also supports event extraction and broader pitch coverage through multiple camera angles where available.

What we learned

  • A useful club product has to match the way teams actually film, not assume broadcast cameras or wearable tracking data.
  • Performance measures become more credible when coaches can trace them back to visible events and player movement.

Have a similar system to build?

Discuss a project