Vision ProTecT

• 

  Start

• 

  Home

• 

  Tutorial

🚀 Project Story: Vision Protect 🌟 Inspiration We live in a hyper-digital era where people — especially Gen Z and students — are glued to their screens. But constant exposure to mobile screens from dangerously close distances can cause eye strain and long-term vision problems. Vision Protect was born from the need to solve this simple yet deeply overlooked problem: How close is too close to your phone?

The idea sparked during late-night coding sessions when we noticed how much time we spent peering into screens, sometimes just inches away. We wanted a minimal, intuitive solution that actively protects users in real-time without disrupting their phone experience.

🛠️ What We Built Vision Protect is an Android application that uses on-device face detection to automatically detect how close a user’s face is to the phone. If the user’s face is too close, the app dims the screen and overlays a semi-transparent warning — a subtle but effective reminder to maintain a healthy viewing distance.

We used:

Kotlin with CameraX API for real-time camera processing

ML Kit's Face Detection from Google to track facial distance

Custom overlay UI components for an elegant UX

Minimal permissions and lightweight performance design

Fully offline processing — your face data stays private

📚 What We Learned Mastered real-time camera input handling using CameraX

Understood the nuances of ML Kit’s face detection pipeline

Learned how to design accessible UI overlays that don’t disrupt user interaction

Gained experience with Android permission handling, app lifecycle events, and thread safety

⚔️ Challenges We Faced Real-time performance: Making sure face detection runs efficiently without draining the battery or causing lag.

Camera permission handling: Ensuring smooth fallback if permissions were denied and providing a clean exit strategy.

UI layering: Getting overlays to work properly across devices and screen sizes was harder than expected.

Face proximity calibration: Tuning the detection threshold (face box width) to trigger warnings accurately across different users and phone cameras.

🎯 What's Next Add customization settings (e.g., proximity sensitivity, dark mode)

Launch on Google Play Store

Build out usage analytics (privacy-focused) to help users track screen hygiene

Explore extending this concept into AR glasses and wearables

👑 Final Words Vision Protect is just the beginning. It's a small app with a big goal: to protect your eyes in the digital age. Designed to be invisible until it matters, it's a health-tech tool for everyone who stares at screens too long — aka all of us.

Your vision is worth protecting. Let your phone help.

📜 Copyright © 2025 Avanish Kasar. All rights reserved.

Built With

• activitycompat
• android
• api)
• camerax
• concise
• contextcompat
• detection
• executorservice
• face
• google
• inputimage
• jetpack
• kit
• kotlin
• ml
• permissions
• runtime
• sdk
• studio
• system
• view