CloPinz Golf — AR Distance Measurement for Golf

The Modern Way to Run CTP and LD Contests

CloPinz is the modern way to run Closest-to-the-Pin (CTP) contests on the golf course — live, accurate, and fully mobile. Whether you’re hosting a tournament or group, managing a golf course, or just playing with friends, CloPinz replaces tape measures, stakes, paper leaderboards, and manual scoring with an accurate, smartphone-based measurement and live leaderboard system. CloPinz is the better and easier way to run your on-course CTP contests. -Create CTP (NTP) contests in under 30 seconds. -Measure directly from your phone’s camera. -Post to the live leaderboard

Industries:

Social

Country:

United States

Services:

Reactive Native, AR Kit Module

Team involved:

Product Designer, Android dev, iOS dev, React Native dev, Backend dev

Duration:

Start in Nov 2025 and ongoing

Technologies

Challenges

Solutions

CTP (short, precision-critical) and LD (long-range, drift-prone) have opposite constraints — yet they run inside the same AR session without resets. What made it hard: - Different error tolerances. - Different spatial scales. - Shared tracking state that cannot be restarted mid-game

Built a shared AR tracking core Swapped measurement strategies at runtime Isolated mode logic while keeping anchors and tracking continuous

AR systems hate long-distance movement — drift compounds fast. World origin drift Tracking degradation outdoors Users walking while holding the device inconsistently

Continuous tracking confidence evaluation Locked critical anchors only after stability thresholds Rejected low-confidence frames instead of guessing

Once the tee is set, it must stay spatially correct — even if ARKit temporarily loses tracking. Device rotation Temporary relocalization Anchor jitter over time

Persistent world-space anchor management Anchor validation before allowing measurement Explicit anchor lifecycle control instead of ARKit defaults

ARKit behaves differently depending on camera, LiDAR, and CPU — yet users expect the same result. Sensor quality variance Performance differences Environmental sensitivity

Device capability detection Adaptive confidence thresholds Normalized output to reduce device-based variance

Users triggering actions too early Mode switches mid-session Unstable tracking states

Hard technical preconditions for every measurement Deterministic gating of the measurement pipeline Explicit failure states instead of silent inaccuracies

Results