WeAreDevelopers LIVE β€’ Apr 28, 2021

Detect Hand Pose with Vision

Milan Todorovic

Want to build touchless interfaces without training custom ML models? Discover how the iOS Vision framework lets you track 21 specific hand landmarks in real-time.

Pause
Mute Enter Fullscreen
#1 about 3 min

Introduction to the Apple Vision framework

The Vision framework provides native support for complex computer vision tasks like image classification and real-time activity detection.

#2 about 3 min

Recognizing hand landmarks and finger joints

The tracking system correctly identifies twenty-one specific coordinates on a human hand including individualized joints and wrist positioning.

#3 about 2 min

Addressing limitations in real-world hand tracking

Environmental edge cases like screen boundaries, irregular hand angles, and physical obstructions can temporarily disrupt accurate camera recognition.

#4 about 3 min

Overview of the sample hand tracking project

The sample tracking application relies on an architectural pattern involving camera views, controllers, and specialized gesture processing files.

#5 about 3 min

Live demonstration of drawing with hand gestures

Running the application on a physical device demonstrates how continuous pinch gestures translate directly into rendering visible screen paths.

#6 about 3 min

Configuring the camera view and request handler

The view controller initializes the camera feed and configures the tracking requests required to process incoming video frame data.

#7 about 7 min

Processing finger tip observations and drawing paths

The application extracts precise landmark coordinates from handler observations and converts them into core graphics points for rendering.

#8 about 3 min

Calculating distance thresholds for pinch detection states

Developers use numeric distance thresholds and state counters to cleanly distinguish between active pinch gestures and resting hand states.

#9 about 5 min

Expanding recognition to human body pose tracking

The underlying mechanisms used for small appendage tracking also apply to mapping structural body joints and overall human posture.

Matching moments

3:56 min

Introduction to text recognition with the Vision framework

Milan Todorovic Β· WWC 2021

4:13 min

Implementing machine learning with Core ML and Vision

MIlan Todorović MIlan Todorović · WWC 2025

3:58 min

Controlling hardware cameras and analyzing visual frames

Marc Rousavy Marc Rousavy Β· WWC 2025

3:43 min

Exploring Unity AR Foundation for advanced tracking logic

Vladimir Novick Β· WWC 2021

2:51 min

Building real-time object tracking with smart glasses

Sasha Denisov Sasha Denisov +1 Β· WWC Europe 2026

3:00 min

Creating intuitive onboarding for controllerless hand tracking

Mario Petricevic Β· WWC 2024

Upcoming sessions on this topic

Open session

World Congress 2026 North America

Point. Ask. Answer. Building Vision into AI Live on Stage.

Kavya Sri Chennoju

Staff AI Engineer at Arm

Kavya Sri Chennoju
Open session

World Congress 2026 North America

From Software Agents to Physical Devices: Inside the Agentic Hardware Stack

Michael Yuan, Vivian Hu

Michael Yuan
Vivian Hu
Open session

World Congress 2026 North America

Vibe Coding Accessibility

Karl Groves

Focused on actively fixing accessibility

Karl Groves
Open session

World Congress 2026 North America

GitHub’s Team X-Ray: Your Repository Knows More About Your Team Than Your Team Does

Andrea Griffiths

Senior Developer Advocate

Andrea Griffiths
Open session

World Congress 2026 North America

Building Stuff with GenAI - The Open Minded Workshop beyond OpenAI

Andreas Erben

CTO for Applied AI and Metaverse at daenet

Andreas Erben
Open session

World Congress 2026 North America

Beyond Vibe Coding: Using CLI Tools as Your AI Counterpart in Pair Programming

Ron Veen

Java enthusiast and Special agent for Team Rockstars IT

Ron Veen