Detecting human actions in a live video feed

At a glance

Project structure

Source bundle/
└── GuessMyExercise/
    ├── App/
    │   └── AppDelegate.swift
    ├── Views/
    │   ├── SummaryViewController.swift
    │   └── MainViewController.swift
    ├── Action Classifier/
    │   ├── ActionPrediction.swift
    │   ├── ExerciseClassifier+Label.swift
    │   └── ExerciseClassifier+FrameRate.swift
    ├── Video Capture/
    │   └── VideoCapture.swift
    ├── Video Processing Chain/
    │   └── VideoProcessingChain.swift
    ├── Pose/
    │   ├── Pose+Connection.swift
    │   ├── Pose+Landmark.swift
    │   └── Pose.swift
    └── Utility/
        └── PerformanceReporter.swift

Structure observations

• Architecturally prominent files are ranked from entry points and role-named declarations; resource-only paths are omitted.
• Primary languages: Swift.
• The verified tree contains 7 project/configuration file(s) and 15 source declaration(s).

Overall architecture

flowchart LR
    N1["AppDelegate"]
    N2["MainViewController"]
    N3["UIKit / AVFoundation APIs"]
    N1 --> N2
    N2 --> N3

Reference code

GuessMyExercise/App/AppDelegate.swift:10 — architecture anchor

@UIApplicationMain
class AppDelegate: UIResponder, UIApplicationDelegate {

    var window: UIWindow?

    func application(_ application: UIApplication,
                     didFinishLaunchingWithOptions launchOptions: [UIApplication.LaunchOptionsKey: Any]?) -> Bool {
        return true
    }
}

Interpretation

The arrows summarize the source-visible entry, role-named types or folders, and framework direction; when nodes come from structural folders, the sequence is a high-level interpretation rather than proof that every adjacent node calls the next. Ownership is claimed only where the next section cites a stored property or assignment. The diagram is intentionally limited to the dominant path into Create ML.

Ownership and state

classDiagram
    AppDelegate o-- UIWindow : window
    SummaryViewController o-- UITableView : tableView
    SummaryViewController o-- Callback : dismissalClosure
    SummaryTableViewCell o-- UILabel : actionLabel

Ownership evidence

GuessMyExercise/App/AppDelegate.swift:13 — stored dependency or nearest verified ownership anchor

@UIApplicationMain
class AppDelegate: UIResponder, UIApplicationDelegate {
    // ...
    var window: UIWindow?
    // ...
}

Composition arrows indicate a source-visible construction expression or locally owned value state; aggregation means the owner stores or receives a dependency without proving exclusive lifetime ownership.

Concurrency, scheduling, and thread safety

Evidence limit: actor isolation, async/await, or Task creation does not by itself prove background-thread execution; Sendable conformance alone does not prove thread-safe mutation.

@MainActor/MainActor.run, DispatchQueue.main, and RunLoop.main are reported as distinct isolation, queue, and event-loop mechanisms. A plain Task is kept separate from Task.detached; neither is labeled as a background thread.

Reference code

GuessMyExercise/Utility/PerformanceReporter.swift:39 — representative execution boundary

class PerformanceReporter {
    // ...
        RunLoop.current.add(timer, forMode: .common)
    // ...
}

State propagation, frameworks, and dependencies

Evidence limit: an import proves a source-level compilation dependency at the cited line; it does not prove runtime use, architectural adoption, or whether a Swift package is a direct application dependency.

receive(on:) describes downstream delivery scheduling, while subscribe(on:) describes upstream subscription/request/cancel scheduling. An import Combine alone establishes neither behavior nor a Store, reducer, Redux, or other application architecture.

Class and protocol design

GuessMyExercise/Video Capture/VideoCapture.swift:16 — representative type boundary

protocol VideoCaptureDelegate: AnyObject {
    /// Informs the delegate when the Video Capture creates a new publisher.
    /// - Parameters:
    ///   - framePublisher: The new frame publisher.
    func videoCapture(_ videoCapture: VideoCapture,
                      didCreate framePublisher: FramePublisher)
}

The source explicitly defines local protocol relationships: MainViewController → VideoCaptureDelegate, MainViewController → VideoProcessingChainDelegate.

Access control

Reference code

GuessMyExercise/Action Classifier/ActionPrediction.swift:63 — representative boundary

    private init(_ otherLabel: AppLabel) {
        label = otherLabel.rawValue
        confidence = nil
    }

Swift declarations without a modifier are internal; explicit private, fileprivate, private(set), public, or open entries above are interpreted by language semantics. Objective-C/C samples instead rely on header and implementation boundaries, which are not equivalent to Swift lexical privacy.

Logic ownership and placement

Design patterns

Naming conventions

• Types: Controller: MainViewController, SummaryViewController; Delegate: AppDelegate, VideoCaptureDelegate, VideoProcessingChainDelegate.
• Protocols: VideoCaptureDelegate, VideoProcessingChainDelegate.
• Methods: application, viewDidLoad, viewDidDisappear, tableView, viewDidAppear, viewWillTransition, onCameraButtonTapped, onSummaryButtonTapped.
• Files: GuessMyExercise/App/AppDelegate.swift, GuessMyExercise/Views/SummaryViewController.swift, GuessMyExercise/Views/MainViewController.swift, GuessMyExercise/Action Classifier/ActionPrediction.swift, GuessMyExercise/Video Capture/VideoCapture.swift, GuessMyExercise/Video Processing Chain/VideoProcessingChain.swift.

Architecture takeaways

• AppDelegate is the main source-visible entry or composition anchor for this sample.
• Framework work reaches UIKit, AVFoundation, CoreML, Vision through a deliberately small high-level chain; the detailed API graph remains inside the cited implementation files.
• Stored-property evidence identifies lifecycle collaboration; it does not by itself prove exclusive object ownership.
• Access-control conclusions separate verified language visibility from the likely design rationale.
• Local protocol relationships provide an explicit substitution boundary.

Source map

• Official Apple documentation