Real-Time Pitch Enforcement
Autonomous Pitch Clock Enforcement
Computer vision that watches the game so you don't have to. Real-time detection. Zero human error.
Demo video coming soon
See it in action
Real college baseball footage. Real CNN-detected pitch events. No simulation — this is the actual model running on three different camera angles.
The Problem
Manual pitch clocks don't scale
The NCAA introduced pitch clocks to speed up the game. But enforcing them still relies on a human operator watching every pitch, every game. That's a problem.
Inconsistent enforcement
Manual operators make different calls game to game. Crews vary. Standards drift.
Human error under pressure
Late innings, close games, hostile crowds. Operators lose focus when it matters most.
Expensive to staff
Every game needs a dedicated operator. Across a full conference schedule, costs add up fast.
How It Works
From camera to clock in under 300ms
Camera Feed
Standard cameras capture the game from multiple angles. No specialized hardware required.
AI Detection
Our computer vision pipeline detects pitch delivery, ball receipt, and game state in real-time.
Clock Signal
Automated signals drive the pitch clock with sub-second precision. No operator needed.
Capabilities
Built for the field, not the lab
Multi-Camera Fusion
Multiple angles feed a single detection pipeline. Redundancy eliminates blind spots.
Edge Deployment
Runs entirely on-site. No cloud dependency during games. No internet required.
Real-Time Processing
Sub-300ms latency from pitch to signal. Fast enough for live game enforcement.
NCAA Ready
Built specifically for college baseball rule enforcement. Conference-level deployment.
Status
In active development
We have working detection models trained on real college baseball game footage. Active testing across multiple camera angles and venues.
Hard launch date coming 2027.
Get notified when we launch