Now in early access

Match footage,
decoded.

Pitch2D turns broadcast video into player and ball tracking data. Calibrate four points, pick two team colours, and let computer vision do the rest.

MP4 · MOV · WebM YOLOv8 + ByteTrack JSON export
TRACKING · 22 OBJECTS
// How it works

Three steps. About a minute.

No setup ceremony. No exhaustive labelling. Just calibrate, configure, and process.

Step 01 — Upload

Drop your footage

MP4, MOV or WebM up to 500 MB. Broadcast TV, drone, or fixed-tripod — all supported.

Step 02 — Calibrate

Four points, four clicks

Click any four known positions on the pitch (corners, box intersections, halfway markers). Pitch2D solves the homography from there.

Step 03 — Process

Get your data

Annotated overlay video, 2D minimap, and frame-by-frame tracking JSON. Ready to drop into Python, R or your own dashboards.

// Privacy

Your footage
stays on your machine.

The Pitch2D engine runs on your local hardware — never on someone else's cloud. Your unreleased match analysis is never uploaded, logged or shared.

  • 01 100% on-device processing — nothing leaves your network.
  • 02 No telemetry, no analytics, no third-party trackers.
  • 03 Air-gap friendly — the backend works fully offline.
  • 04 You own every byte of output: video, JSON, the lot.
// pipeline local
input match.mp4
processor YOLOv8 · GPU
network egress none ✓
output overlay.mp4 · tracking.json
// What you get

The full toolkit.

Everything you need to turn a video file into structured tactical data.

Player & ball detection

YOLOv8 pre-trained on COCO, paired with ByteTrack to keep IDs stable across cuts and occlusions.

4-point homography

Click-to-calibrate transform from camera view to top-down pitch coordinates. Survives camera pans.

Auto team colours

HSV clustering on the torso region with per-track-ID lock-in. Referees auto-detected and labelled.

Smart shot filter

Replays, close-ups and studio frames are detected and skipped — your tracking data stays clean.

Structured export

Per-frame JSON: pixel coordinates, pitch coordinates, team IDs, track IDs. Pandas-ready.

GPU accelerated

Auto-detects CUDA. ~3-5× real-time on a modern NVIDIA card; CPU fallback for everything else.

For analysts

Drop-in tracking data

Skip manual annotation. Get pitch-coordinate data into your notebook in minutes, not days.

For scouts

Quantify the eye test

Heat maps, average positions, off-ball runs — measured straight from any clip you can find.

For clubs

Self-hosted, private

No video leaves your facility. Run it on a single workstation behind your firewall.

// Ready when you are

Stop watching tape.
Start querying it.

One video file is all it takes.

Open the app