World Congress 2021 • Jun 30, 2021

Let your iOS app read texts

Milan Todorovic

Add offline OCR to your iOS app in just a few lines of code. Discover how Swift's Vision framework extracts real-world text while completely preserving user privacy.

Pause
Mute Enter Fullscreen
#1 about 4 min

Introduction to text recognition with the Vision framework

Integrating modern optical character recognition capabilities into native iOS applications using Swift.

#2 about 5 min

Processing native image data with Vision framework request handlers

Handling document pipelines by connecting camera snapshots and image galleries to algorithmic request handlers.

#3 about 3 min

Implementing recognized text requests and handling candidate outputs

Generating specific code commands that return structured observation arrays for categorizing readable string alternatives.

#4 about 4 min

Trade-offs between fast and accurate text recognition levels

Balancing immediate asynchronous performance metrics with processor-heavy deep learning neural networks.

#5 about 4 min

Configuring custom lexicons for domain-specific text recognition accuracy

Improving string reading probability and logic by passing domain custom vocabularies into language transcription correctors.

#6 about 9 min

Demonstrating text recognition from books, business cards, and receipts

Seeing the framework independently capture real world variables across printed pages, contact cards, and bilingual sales receipts.

#7 about 2 min

Extracting characters from organic and messy handwritten text inputs

Testing the natural boundaries of digital optical extraction models against imperfectly shaped unstructured human lettering.

#8 about 3 min

Reviewing Xcode implementations and Vision framework documentation reference resources

Exploring Swift file components and navigating the Apple developer ecosystem libraries to discover and extend algorithmic features.

#9 about 7 min

Evaluating on-device data privacy, model performance, and cross-platform alternatives

Discussing the inherent security of zero network architecture and contrasting internal Vision processing speeds against Google light models.

Matching moments

2:21 min

Introduction to the Apple Vision framework

Milan Todorovic · LIVE

3:56 min

Using Mac OCR for offline image text extraction

Chris Heilmann Chris Heilmann +2 · LIVE

1:22 min

Extracting text from images using the Mac OCR tool

Chris Heilmann Chris Heilmann +1 · LIVE

4:13 min

Implementing machine learning with Core ML and Vision

MIlan Todorović MIlan Todorović · World Congress 2025

2:25 min

Testing the AI generated Apple iOS Flashcards application

MIlan Todorović MIlan Todorović · World Congress 2026 Europe

4:47 min

Relying on artificial intelligence for captions and environment mapping

Chris Heilmann Chris Heilmann +2 · LIVE