DetectorKit

Easy object detection, framing and obfuscation with Swift

Architecture

The light gray boxes at the top of the architecture diagram above are the public interface into DetectorKit. Through these you can easily access the core features of detecting, framing and obfuscating faces and text within images.

Usage

Detecting

Faces

// Detect all faces within an image
let faceDetector = FaceDetector(image)

// Then you can easily access the faces (as CIFeatures)
let faces = faceDetector.faces

// Access underlying CIDetector
let detector = faceDetector.detector

Framing

Faces

let faceFramer = FaceFramer(imageView)

// Change color and borderWidth 
// Note: borderWidth only works for box() and radial()
faceFramer.shapeColor = UIColor.red
faceFramer.borderWidth = 2.0

// Frame faces with a rectangle
faceFramer.box()

// or with a radial frame
faceFramer.radial()

// or with a pulsing circle
faceFramer.pulse()

// or with rotating boxes
faceFramer.boxRotate()

Obfuscating

Faces

let faceObscure = FaceObscure(image)

// pixelate all faces within an image
faceObscure.pixelate()

// or use a crystalize filter
faceObscure.crystalize()

Installation

The easiest way the use DetectorKit is with CocoaPods

Installation with CocoaPods

To integrate DetectorKit into your Xcode project using CocoaPods, specify it in your Podfile

platform :ios, '10.0'

target 'TargetName' do
  pod 'DetectorKit'
end

Then, run the following command:

$ pod install

Example

For an example app implementing the above usage, take a look at Sherlock

License

DetectorKit is released under the MIT license. See LICENSE for details.