mobileai-iot

Palmistry

TensorFlow.js & MediaPipe Hands Edge AI Computer Vision App

Real-time mobile app capturing hand images to detect palm lines using on-device TensorFlow.js and MediaPipe computer vision models.

Client / Industry
Entertainment AI Startup
My Role
Mobile AI Specialist
Engineering Timeline
2 Months
Core Technologies & Architecture Stack
React NativeTensorFlow.jsMediaPipe HandsVision CameraMMKV
🚀 Production Architecture Case Study

Palmistry

TensorFlow.js & MediaPipe Hands Edge AI Computer Vision App

01. The Challenge & Problem

What Needed to Be Solved

An entertainment app required real-time camera hand detection and on-device machine learning inference without sending raw photos to external cloud servers.

02. System Architecture

How I Engineered the Solution

Integrated TensorFlow.js, Fast TFLite, and MediaPipe Hands with react-native-vision-camera for instant 30+ FPS landmark tracking, MMKV local cache, and AdMob monetization.

03. Core Capabilities & Features

Key Production Features Built

On-device TensorFlow.js & MediaPipe hand-landmark computer vision processing
Real-time high frame rate camera capture via Vision Camera
AdMob rewarded video ads and interstitial revenue integrations
MMKV ultra-fast local storage running 30x faster than standard storage
04. Production Result & Business Impact

Sustained 30+ FPS real-time hand tracking directly on mobile hardware with full privacy protection.

Build Similar Project
⚡ Direct Senior Engineer • Fast Turnaround • 100% IP Ownership

Need an App or Platform Like Palmistry?

Work directly with Inayat Ali (Founder & Senior Full-Stack Architect) to design, build, launch, or scale your mobile app and web platform—with transparent milestones and zero agency middlemen.

Chat on WhatsApp Now Send Project Scope