main

Source: main.js:2

Instance Methods

loadModels()

Loads all required face-api.js models for facial detection, landmarks, recognition, and age/gender. This is called during application startup to ensure the models are cached before any webcam processing.

• See:
  • init – the main initialization routine that invokes `loadModels` before starting the camera.
  • setStatus – updates UI status while models are being loaded.

handleError(err: Error, fallbackMessage: string)

Centralized error handling for UI actions. Logs the fallback message, the error details, updates the status UI, shows the placeholder, and records the full message in the log.

Parameters

• err (Error) — The caught error object.
• fallbackMessage (string) — User‑friendly message describing the context of the error.

• See:
  • init – camera-start failures are routed here.
  • startCamera – camera-switch failures are routed here.

buildFaceapiSaveSection(saved2d: Object): object

Build the faceapi section for a save action. After saving, the best match IS the saved face (dist ≈ 0), so we report detectionState: 'matched' unconditionally.

Parameters

• saved2d (Object)

Returns

• object

buildMediapipeSaveSection(saved3d: Object, id: number): object

Build the mediapipe section for a save action.

Parameters

• saved3d (Object)
• id (number)

Returns

• object

Static Methods

setBusy(isBusy: boolean)

Sets the busy flag for the whole UI, disabling/enabling controls during asynchronous operations.

Parameters

• isBusy (boolean) — When true, UI controls are disabled to prevent concurrent actions.

• See:
  • init – called during startup to manage UI state while models load.
  • loadModels – UI is set busy while models are loading.
  • saveFace – save actions hold the UI busy while both engines persist data.
  • openAnalyzePanel – analysis actions use the same guard against concurrent work.

computeOverall( faceapiSection: object | null, mediapipeSection: object | null, ): 'matched' | 'eluded' | 'partial-elusion' | 'unknown'

Compute the overall field of a matchStateChanged payload.

Parameters

• faceapiSection (object | null)
• mediapipeSection (object | null)

Returns

• 'matched' | 'eluded' | 'partial-elusion' | 'unknown'

init()

Initializes the application: loads the database, renders statistics, sets up the canvas, registers UI event listeners, loads models, ghostyle plugins, and starts the webcam. In production this is the module entry point; Vitest skips the automatic call so individual helpers can be isolated.

Fires

• window#gstmxxReady after models, plugins,event: and the camera are ready.

• See:
  • loadModels – called within init to load face-api.js models before webcam activation.