Vision Framework API Reference

Comprehensive reference for Vision framework people-focused computer vision: subject segmentation, hand/body pose detection, person detection, and face analysis.

When to Use This Reference

• Implementing subject lifting using VisionKit or Vision
• Detecting hand/body poses for gesture recognition or fitness apps
• Segmenting people from backgrounds or separating multiple individuals
• Face detection and landmarks for AR effects or authentication
• Combining Vision APIs to solve complex computer vision problems
• Looking up specific API signatures and parameter meanings

Related skills: See vision for decision trees and patterns, vision-diag for troubleshooting

Vision Framework Overview

Vision provides computer vision algorithms for still images and video:

Core workflow:

1. Create request (e.g., VNDetectHumanHandPoseRequest())
2. Create handler with image (VNImageRequestHandler(cgImage: image))
3. Perform request (try handler.perform([request]))
4. Access observations from request.results

Coordinate system: Lower-left origin, normalized (0.0-1.0) coordinates

Performance: Run on background queue - resource intensive, blocks UI if on main thread

Subject Segmentation APIs

VNGenerateForegroundInstanceMaskRequest

Availability: iOS 17+, macOS 14+, tvOS 17+, visionOS 1+

Generates class-agnostic instance mask of foreground objects (people, pets, buildings, food, shoes, etc.)

Basic Usage

let request = VNGenerateForegroundInstanceMaskRequest()
let handler = VNImageRequestHandler(cgImage: image)

try handler.perform([request])

guard let observation = request.results?.first as? VNInstanceMaskObservation else {
    return
}

InstanceMaskObservation

allInstances: IndexSet containing all foreground instance indices (excludes background 0)

instanceMask: CVPixelBuffer with UInt8 labels (0 = background, 1+ = instance indices)

instanceAtPoint(_:): Returns instance index at normalized point

let point = CGPoint(x: 0.5, y: 0.5)  // Center of image
let instance = observation.instanceAtPoint(point)

if instance == 0 {
    print("Background tapped")
} else {
    print("Instance \(instance) tapped")
}

Generating Masks

createScaledMask(for:croppedToInstancesContent:)

Parameters:

• for: IndexSet of instances to include
• croppedToInstancesContent:
  • false = Output matches input resolution (for compositing)
  • true = Tight crop around selected instances

Returns: Single-channel floating-point CVPixelBuffer (soft segmentation mask)

// All instances, full resolution
let mask = try observation.createScaledMask(
    for: observation.allInstances,
    croppedToInstancesContent: false
)

// Single instance, cropped
let instances = IndexSet(integer: 1)
let croppedMask = try observation.createScaledMask(
    for: instances,
    croppedToInstancesContent: true
)

Instance Mask Hit Testing

Access raw pixel buffer to map tap coordinates to instance labels:

let instanceMask = observation.instanceMask

CVPixelBufferLockBaseAddress(instanceMask, .readOnly)
defer { CVPixelBufferUnlockBaseAddress(instanceMask, .readOnly) }

let baseAddress = CVPixelBufferGetBaseAddress(instanceMask)
let width = CVPixelBufferGetWidth(instanceMask)
let bytesPerRow = CVPixelBufferGetBytesPerRow(instanceMask)

// Convert normalized tap to pixel coordinates
let pixelPoint = VNImagePointForNormalizedPoint(
    CGPoint(x: normalizedX, y: normalizedY),
    width: imageWidth,
    height: imageHeight
)

// Calculate byte offset
let offset = Int(pixelPoint.y) * bytesPerRow + Int(pixelPoint.x)

// Read instance label
let label = UnsafeRawPointer(baseAddress!).load(
    fromByteOffset: offset,
    as: UInt8.self
)

let instances = label == 0 ? observation.allInstances : IndexSet(integer: Int(label))

VisionKit Subject Lifting

ImageAnalysisInteraction (iOS)

Availability: iOS 16+, iPadOS 16+

Adds system-like subject lifting UI to views:

let interaction = ImageAnalysisInteraction()
interaction.preferredInteractionTypes = .imageSubject  // Or .automatic
imageView.addInteraction(interaction)

Interaction types:

• .automatic: Subject lifting + Live Text + data detectors
• .imageSubject: Subject lifting only (no interactive text)

ImageAnalysisOverlayView (macOS)

Availability: macOS 13+

let overlayView = ImageAnalysisOverlayView()
overlayView.preferredInteractionTypes = .imageSubject
nsView.addSubview(overlayView)

Programmatic Access

ImageAnalyzer

let analyzer = ImageAnalyzer()
let configuration = ImageAnalyzer.Configuration([.text, .visualLookUp])

let analysis = try await analyzer.analyze(image, configuration: configuration)

ImageAnalysis

subjects: [Subject] - All subjects in image

highlightedSubjects: Set<Subject> - Currently highlighted (user long-pressed)

subject(at:): Async lookup of subject at normalized point (returns nil if none)

// Get all subjects
let subjects = analysis.subjects

// Look up subject at tap
if let subject = try await analysis.subject(at: tapPoint) {
    // Process subject
}

// Change highlight state
analysis.highlightedSubjects = Set([subjects[0], subjects[1]])

Subject Struct

image: UIImage/NSImage - Extracted subject with transparency

bounds: CGRect - Subject boundaries in image coordinates

// Single subject image
let subjectImage = subject.image

// Composite multiple subjects
let compositeImage = try await analysis.image(for: [subject1, subject2])

Out-of-process: VisionKit analysis happens out-of-process (performance benefit, image size limited)

Person Segmentation APIs

VNGeneratePersonSegmentationRequest

Availability: iOS 15+, macOS 12+

Returns single mask containing all people in image:

let request = VNGeneratePersonSegmentationRequest()
// Configure quality level if needed
try handler.perform([request])

guard let observation = request.results?.first as? VNPixelBufferObservation else {
    return
}

let personMask = observation.pixelBuffer  // CVPixelBuffer

VNGeneratePersonInstanceMaskRequest

Availability: iOS 17+, macOS 14+

Returns separate masks for up to 4 people:

let request = VNGeneratePersonInstanceMaskRequest()
try handler.perform([request])

guard let observation = request.results?.first as? VNInstanceMaskObservation else {
    return
}

// Same InstanceMaskObservation API as foreground instance masks
let allPeople = observation.allInstances  // Up to 4 people (1-4)

// Get mask for person 1
let person1Mask = try observation.createScaledMask(
    for: IndexSet(integer: 1),
    croppedToInstancesContent: false
)

Limitations:

• Segments up to 4 people
• With >4 people: may miss people or combine them (typically background people)
• Use VNDetectFaceRectanglesRequest to count faces if you need to handle crowded scenes

Hand Pose Detection

VNDetectHumanHandPoseRequest

Availability: iOS 14+, macOS 11+

Detects 21 hand landmarks per hand:

let request = VNDetectHumanHandPoseRequest()
request.maximumHandCount = 2  // Default: 2, increase if needed

let handler = VNImageRequestHandler(cgImage: image)
try handler.perform([request])

for observation in request.results as? [VNHumanHandPoseObservation] ?? [] {
    // Process each hand
}

Performance note: maximumHandCount affects latency. Pose computed only for hands ≤ maximum. Set to lowest acceptable value.

Hand Landmarks (21 points)

Wrist: 1 landmark

Thumb (4 landmarks):

• .thumbTip
• .thumbIP (interphalangeal joint)
• .thumbMP (metacarpophalangeal joint)
• .thumbCMC (carpometacarpal joint)

Fingers (4 landmarks each):

• Tip (.indexTip, .middleTip, .ringTip, .littleTip)
• DIP (distal interphalangeal joint)
• PIP (proximal interphalangeal joint)
• MCP (metacarpophalangeal joint)

Group Keys

Access landmark groups:

Group Key	Points
.all	All 21 landmarks
.thumb	4 thumb joints
.indexFinger	4 index finger joints
.middleFinger	4 middle finger joints
.ringFinger	4 ring finger joints
.littleFinger	4 little finger joints

// Get all points
let allPoints = try observation.recognizedPoints(.all)

// Get index finger points only
let indexPoints = try observation.recognizedPoints(.indexFinger)

// Get specific point
let thumbTip = try observation.recognizedPoint(.thumbTip)
let indexTip = try observation.recognizedPoint(.indexTip)

// Check confidence
guard thumbTip.confidence > 0.5 else { return }

// Access location (normalized coordinates, lower-left origin)
let location = thumbTip.location  // CGPoint

Gesture Recognition Example (Pinch)

let thumbTip = try observation.recognizedPoint(.thumbTip)
let indexTip = try observation.recognizedPoint(.indexTip)

guard thumbTip.confidence > 0.5, indexTip.confidence > 0.5 else {
    return
}

let distance = hypot(
    thumbTip.location.x - indexTip.location.x,
    thumbTip.location.y - indexTip.location.y
)

let isPinching = distance < 0.05  // Normalized threshold

Chirality (Handedness)

let chirality = observation.chirality  // .left or .right or .unknown

Body Pose Detection

VNDetectHumanBodyPoseRequest (2D)

Availability: iOS 14+, macOS 11+

Detects 18 body landmarks (2D normalized coordinates):

let request = VNDetectHumanBodyPoseRequest()
try handler.perform([request])

for observation in request.results as? [VNHumanBodyPoseObservation] ?? [] {
    // Process each person
}

Body Landmarks (18 points)

Face (5 landmarks):

• .nose, .leftEye, .rightEye, .leftEar, .rightEar

Arms (6 landmarks):

• Left: .leftShoulder, .leftElbow, .leftWrist
• Right: .rightShoulder, .rightElbow, .rightWrist

Torso (7 landmarks):

• .neck (between shoulders)
• .leftShoulder, .rightShoulder (also in arm groups)
• .leftHip, .rightHip
• .root (between hips)

Legs (6 landmarks):

• Left: .leftHip, .leftKnee, .leftAnkle
• Right: .rightHip, .rightKnee, .rightAnkle

Note: Shoulders and hips appear in multiple groups

Group Keys (Body)

Group Key	Points
.all	All 18 landmarks
.face	5 face landmarks
.leftArm	shoulder, elbow, wrist
.rightArm	shoulder, elbow, wrist
.torso	neck, shoulders, hips, root
.leftLeg	hip, knee, ankle
.rightLeg	hip, knee, ankle

// Get all body points
let allPoints = try observation.recognizedPoints(.all)

// Get left arm only
let leftArmPoints = try observation.recognizedPoints(.leftArm)

// Get specific joint
let leftWrist = try observation.recognizedPoint(.leftWrist)

VNDetectHumanBodyPose3DRequest (3D)

Availability: iOS 17+, macOS 14+

Returns 3D skeleton with 17 joints in meters (real-world coordinates):

let request = VNDetectHumanBodyPose3DRequest()
try handler.perform([request])

guard let observation = request.results?.first as? VNHumanBodyPose3DObservation else {
    return
}

// Get 3D joint position
let leftWrist = try observation.recognizedPoint(.leftWrist)
let position = leftWrist.position  // simd_float4x4 matrix
let localPosition = leftWrist.localPosition  // Relative to parent joint

3D Body Landmarks (17 points): Same as 2D except no ears (15 vs 18 2D landmarks)

3D Observation Properties

bodyHeight: Estimated height in meters

• With depth data: Measured height
• Without depth data: Reference height (1.8m)

heightEstimation: .measured or .reference

cameraOriginMatrix: simd_float4x4 camera position/orientation relative to subject

pointInImage(_:): Project 3D joint back to 2D image coordinates

let wrist2D = try observation.pointInImage(leftWrist)

3D Point Classes

VNPoint3D: Base class with simd_float4x4 position matrix

VNRecognizedPoint3D: Adds identifier (joint name)

VNHumanBodyRecognizedPoint3D: Adds localPosition and parentJoint

// Position relative to skeleton root (center of hip)
let modelPosition = leftWrist.position

// Position relative to parent joint (left elbow)
let relativePosition = leftWrist.localPosition

Depth Input

Vision accepts depth data alongside images:

// From AVDepthData
let handler = VNImageRequestHandler(
    cvPixelBuffer: imageBuffer,
    depthData: depthData,
    orientation: orientation
)

// From file (automatic depth extraction)
let handler = VNImageRequestHandler(url: imageURL)  // Depth auto-fetched

Depth formats: Disparity or Depth (interchangeable via AVFoundation)

LiDAR: Use in live capture sessions for accurate scale/measurement

Face Detection & Landmarks

VNDetectFaceRectanglesRequest

Availability: iOS 11+

Detects face bounding boxes:

let request = VNDetectFaceRectanglesRequest()
try handler.perform([request])

for observation in request.results as? [VNFaceObservation] ?? [] {
    let faceBounds = observation.boundingBox  // Normalized rect
}

VNDetectFaceLandmarksRequest

Availability: iOS 11+

Detects face with detailed landmarks:

let request = VNDetectFaceLandmarksRequest()
try handler.perform([request])

for observation in request.results as? [VNFaceObservation] ?? [] {
    if let landmarks = observation.landmarks {
        let leftEye = landmarks.leftEye
        let nose = landmarks.nose
        let leftPupil = landmarks.leftPupil  // Revision 2+
    }
}

Revisions:

• Revision 1: Basic landmarks
• Revision 2: Detects upside-down faces
• Revision 3+: Pupil locations

Person Detection

VNDetectHumanRectanglesRequest

Availability: iOS 13+

Detects human bounding boxes (torso detection):

let request = VNDetectHumanRectanglesRequest()
try handler.perform([request])

for observation in request.results as? [VNHumanObservation] ?? [] {
    let humanBounds = observation.boundingBox  // Normalized rect
}

Use case: Faster than pose detection when you only need location

CoreImage Integration

CIBlendWithMask Filter

Composite subject on new background using Vision mask:

// 1. Get mask from Vision
let observation = request.results?.first as? VNInstanceMaskObservation
let visionMask = try observation.createScaledMask(
    for: observation.allInstances,
    croppedToInstancesContent: false
)

// 2. Convert to CIImage
let maskImage = CIImage(cvPixelBuffer: visionMask)

// 3. Apply filter
let filter = CIFilter(name: "CIBlendWithMask")!
filter.setValue(sourceImage, forKey: kCIInputImageKey)
filter.setValue(maskImage, forKey: kCIInputMaskImageKey)
filter.setValue(newBackground, forKey: kCIInputBackgroundImageKey)

let output = filter.outputImage  // Composited result

Parameters:

• Input image: Original image to mask
• Mask image: Vision's soft segmentation mask
• Background image: New background (or empty image for transparency)

HDR preservation: CoreImage preserves high dynamic range from input (Vision/VisionKit output is SDR)

API Quick Reference

Subject Segmentation

API	Platform	Purpose
VNGenerateForegroundInstanceMaskRequest	iOS 17+	Class-agnostic subject instances
VNGeneratePersonInstanceMaskRequest	iOS 17+	Up to 4 people separately
VNGeneratePersonSegmentationRequest	iOS 15+	All people (single mask)
ImageAnalysisInteraction (VisionKit)	iOS 16+	UI for subject lifting

Pose Detection

API	Platform	Landmarks	Coordinates
VNDetectHumanHandPoseRequest	iOS 14+	21 per hand	2D normalized
VNDetectHumanBodyPoseRequest	iOS 14+	18 body joints	2D normalized
VNDetectHumanBodyPose3DRequest	iOS 17+	17 body joints	3D meters

Face & Person Detection

API	Platform	Purpose
VNDetectFaceRectanglesRequest	iOS 11+	Face bounding boxes
VNDetectFaceLandmarksRequest	iOS 11+	Face with detailed landmarks
VNDetectHumanRectanglesRequest	iOS 13+	Human torso bounding boxes

Observation Types

Observation	Returned By
VNInstanceMaskObservation	Foreground/person instance masks
VNPixelBufferObservation	Person segmentation (single mask)
VNHumanHandPoseObservation	Hand pose
VNHumanBodyPoseObservation	Body pose (2D)
VNHumanBodyPose3DObservation	Body pose (3D)
VNFaceObservation	Face detection/landmarks
VNHumanObservation	Human rectangles

WWDC Sessions

WWDC 2023:

• 10176: Lift subjects from images in your app
• 111241: Explore 3D body pose and person segmentation in Vision
• 10048: What's new in VisionKit

WWDC 2022:

• 10024: What's new in Vision

WWDC 2020:

• 10653: Detect Body and Hand Pose with Vision
• 10043: Build an Action Classifier with Create ML
• 10099: Explore the Action & Vision app

Apple Documentation

• Vision Framework
• VisionKit Framework
• Detecting Hand Poses with Vision
• CoreImage Filter Reference

Related Axiom Skills

• vision — Decision trees, patterns, anti-patterns
• vision-diag — Troubleshooting when things go wrong