We 're justtrackingpoints

and that’s all

We are not a $5k hardware trap. We do not drill kiosks into the court. We just turn standard 1080p 60fps IP cameras into an automated AI referee & live match dashboard at the net.

What does it mean Tapy?

Tapy is an automated AI referee and live match dashboard running on an on-court tablet. No sensors in rackets. No $5k hardware trap. Just a standard 1080p 60 FPS IP camera mounted on the fence, an edge computer vision pipeline, and a net-side tablet managing the match in real time.

Purpose of using Tapy

The problem

Padel is blowing up worldwide, but actual matches on court are pure chaos and endless ego disputes:

Score Amnesia: Three long rallies in and nobody has any clue whether it's 30–15 or 40–30. Players burn two minutes yelling across the net. Pure delusion.

Cage vs. Glass Beef: The ball grazes the back wall and drama starts instantly: “Bro, that was 100% glass, no cap!” vs. “Are you blind? That was straight cage!” Matches regularly end on the brink of physical violence.

Hardware Gatekeeping: Legacy solutions like Wingfield or PlaySight demand $3,000 to $6,000+ for massive kiosks, complex wiring, and proprietary sensors. Local club owners want nothing to do with that.

Solution

Tapy turns standard 1080p 60 FPS cameras into an automated AI referee and live match dashboard on an on-court tablet:

Zero sensors in rackets. Zero proprietary balls. The moment the ball bounces twice or sails out, the score flips automatically in under a second:

  • Real-time Scoreboard (30 — 15)

    What’sApp

  • 10s VAR Rewind (Beef Canceller)

    Email

  • 2D Court Radar (Minimap)

    Telegram

  • Smash Velocity Radar (112 km/h)

    Facebook Messanger

  • Automated Clout Clips (+1000 Aura)

    Discord

  • Zero Hardware Gatekeeping

    Etc.

Designed to make life {easier}

It is enough to waste court time on toxic arguments and forgotten scores. We know that whenever you're on court, the least you want to do is argue over cage vs. glass or shout the score after every single point.

H o w   i t ’ s   c o o k e d   u n d e r   t h e   h o o d ?

  • Step 1

    1080p 60 FPS RTSP Court Camera Feed — Mount any standard camera on the back fence

  • Step 2

    OpenCV Homography Calibration — 4 clicks map camera pixels to real 10x20m metric court coordinates

  • Step 3

    Pre-trained YOLOv8-Pose & TrackNet — Track player skeletons, ball trajectory & bounce frames at 60 FPS

  • Step 4

    Python Rules Engine & WebSockets — Evaluates FIP rules and streams score JSON (<50ms) to the net tablet

Integration of your form with different {messengers}

Most often, the data that a potential client leaves in the form are sent to email or to the CRM system. But the fact is that a small business does not always use a CRM system, and also does not always have a manager who monitors customer requests via email in real time

 
A few important benefits
 
A few important benefits
 
A few important benefits
 
A few important benefits

Each of these features has been engineered specifically for padel players and club owners who want zero drama, automated scoring, and instant social clout.

  • Zero sensors in rackets. Zero proprietary balls. Standard 1080p 60 FPS camera mounted on the fence does all the heavy lifting via computer vision.

    Zero Sensor Traps

  • The most crucial button in the app. Hit the big red button at the net and the last 10 seconds instantly loop in slow motion. Ego checked, court beef canceled.

    10s VAR Rewind (Beef Canceller)

  • Automatic score progression (0 -> 15 -> 30 -> 40 -> Game) updated in under a second when the ball bounces twice or sails out. No more yelling across the net.

    Instant Scoreboard

  • Top-down tactical radar view just like FIFA or CS:GO. Four glowing dots live in metric court space show who commanded the net vs. who camped at the glass.

    2D Court Radar (Minimap)

  • Clock smash velocity in real time on the tablet (112 km/h Vibora). Real-time telemetry that validates your power and gives you instant bragging rights.

    Smash Speed Radar

  • Automatically cuts the match's nastiest rallies into vertical 9:16 highlight reels ready for Instagram Stories and TikTok. Pure +1000 aura for your feed.

    Automated Clout Clips (+1000 Aura)

Tapy relieves the headache

of developers and site owners

  • For Players

    Zero score amnesia. No more yelling across the net or arguing whether the ball touched glass or cage. Get instant slow-mo VAR replay at the net and automated highlight clips for your socials.

  • For Club Owners

    Upgrade every court into a high-tech smart court without spending $5,000 on proprietary hardware kiosks. Mount a standard IP camera, connect an iPad at the net, and attract younger, tech-savvy players.

Problems that we solve

Website conversion increases

Instead of showing a boring static chart or CSV, you demo live video with glowing ball trajectories, real-time score flipping, and instant slow-mo VAR replay.

Less time to set up the form

We train zero models from scratch. Pre-trained YOLOv8 and TrackNet models mean nobody is manually drawing bounding boxes around 5,000 padel balls until 4 AM.

You don't need to know how to program

Nails the complete full-stack machine learning package: applied deep learning, geometric computer vision, low-latency WebSockets, and clean product UI.

Project roadmap

Engineered for an on-time, bulletproof Demo Day delivery with zero model retraining:

{Phase 1: Ingestion & Homography}Days 1–3
jan 19

{Phase 2: Vision & Bounce FSM}

Days 4–6

{Phase 3: WebSocket & Tablet UI}

Days 7–8

{Phase 4: VAR Replay & Demo Polish}

FAQ

01

Do we need expensive sensors in the rackets or balls?

Zero sensors. Tapy runs purely on computer vision using a standard 1080p 60 FPS IP camera mounted on the back fence. No chips in balls, no sensor tapes on rackets.
02

How does Tapy resolve cage vs. glass disputes?

Just follow these three steps:
1. Create an copy your personal ID link.
2. OpenCV Homography Calibration — 4 clicks map camera pixels to real 10x20m metric court coordinates.
3. Pre-trained YOLOv8-Pose & TrackNet — Track player skeletons, ball trajectory & bounce frames at 60 FPS.
4. Python Rules Engine & WebSockets — Evaluates FIP rules and streams score JSON (<50ms) to the net tablet.
03

What is the scoreboard update latency?

Under 50 milliseconds! The Python state machine evaluates the rally in real time and broadcasts live score JSON over a local WebSocket stream directly to the tablet at the net.
04

Will this really work reliably for a 10-day sprint?

100%. We train zero models from scratch. By standing on the shoulders of pre-trained YOLOv8-Pose and TrackNet combined with classic geometric computer vision, the pipeline is reproducible, fast, and immune to training failures.