Config | @vladmandic/human

Configuration interface definition for Human library

Contains all configurable parameters

Hierarchy

Config

Index

Properties

async

async: boolean

Perform model loading and inference concurrently or sequentially

backend

backend: null | string

Backend used for TFJS operations

body

body: { enabled: boolean; maxDetected: number; minConfidence: number; modelPath: string; skipFrames: number }

Controlls and configures all body detection specific options

enabled: true/false
modelPath: body pose model, can be absolute path or relative to modelBasePath
minConfidence: threshold for discarding a prediction
maxDetected: maximum number of people detected in the input, should be set to the minimum number for performance

Type declaration

enabled: boolean
maxDetected: number
minConfidence: number
modelPath: string
skipFrames: number

cacheSensitivity

cacheSensitivity: number

Cache sensitivity

values 0..1 where 0.01 means reset cache if input changed more than 1%
set to 0 to disable caching

debug

debug: boolean

Print debug statements to console

face

face: { description: { enabled: boolean; minConfidence: number; modelPath: string; skipFrames: number }; detector: { iouThreshold: number; maxDetected: number; minConfidence: number; modelPath: string; return: boolean; rotation: boolean; skipFrames: number }; emotion: { enabled: boolean; minConfidence: number; modelPath: string; skipFrames: number }; enabled: boolean; iris: { enabled: boolean; modelPath: string }; mesh: { enabled: boolean; modelPath: string } }

Controlls and configures all face-specific options:

face detection, face mesh detection, age, gender, emotion detection and face description Parameters:
enabled: true/false
modelPath: path for each of face models
minConfidence: threshold for discarding a prediction
iouThreshold: ammount of overlap between two detected objects before one object is removed
maxDetected: maximum number of faces detected in the input, should be set to the minimum number for performance
rotation: use calculated rotated face image or just box with rotation as-is, false means higher performance, but incorrect mesh mapping on higher face angles
return: return extracted face as tensor for futher user processing, in which case user is reponsible for manually disposing the tensor

Type declaration

description: { enabled: boolean; minConfidence: number; modelPath: string; skipFrames: number }
- enabled: boolean
- minConfidence: number
- modelPath: string
- skipFrames: number
detector: { iouThreshold: number; maxDetected: number; minConfidence: number; modelPath: string; return: boolean; rotation: boolean; skipFrames: number }
- iouThreshold: number
- maxDetected: number
- minConfidence: number
- modelPath: string
- return: boolean
- rotation: boolean
- skipFrames: number
emotion: { enabled: boolean; minConfidence: number; modelPath: string; skipFrames: number }
- enabled: boolean
- minConfidence: number
- modelPath: string
- skipFrames: number
enabled: boolean
iris: { enabled: boolean; modelPath: string }
- enabled: boolean
- modelPath: string
mesh: { enabled: boolean; modelPath: string }
- enabled: boolean
- modelPath: string

filter

filter: { blur: number; brightness: number; contrast: number; enabled: boolean; flip: boolean; height: number; hue: number; kodachrome: boolean; negative: boolean; pixelate: number; polaroid: boolean; return: boolean; saturation: number; sepia: boolean; sharpness: number; technicolor: boolean; vintage: boolean; width: number }

Run input through image filters before inference

image filters run with near-zero latency as they are executed on the GPU

Type declaration

blur: number

Range: 0 (no blur) to N (blur radius in pixels)
brightness: number

Range: -1 (darken) to 1 (lighten)
contrast: number

Range: -1 (reduce contrast) to 1 (increase contrast)
enabled: boolean
flip: boolean

Flip input as mirror image
height: number
Resize input height
- if both width and height are set to 0, there is no resizing
- if just one is set, second one is scaled automatically
- if both are set, values are used as-is
hue: number

Range: 0 (no change) to 360 (hue rotation in degrees)
kodachrome: boolean

Image kodachrome colors
negative: boolean

Image negative
pixelate: number

Range: 0 (no pixelate) to N (number of pixels to pixelate)
polaroid: boolean

Image polaroid camera effect
return: boolean

Return processed canvas imagedata in result
saturation: number

Range: -1 (reduce saturation) to 1 (increase saturation)
sepia: boolean

Image sepia colors
sharpness: number

Range: 0 (no sharpening) to 1 (maximum sharpening)
technicolor: boolean

Image technicolor colors
vintage: boolean

Image vintage colors
width: number
Resize input width
- if both width and height are set to 0, there is no resizing
- if just one is set, second one is scaled automatically
- if both are set, values are used as-is

gesture

gesture: { enabled: boolean }

Controlls gesture detection

Type declaration

enabled: boolean

hand

hand: { detector: { modelPath: string }; enabled: boolean; iouThreshold: number; landmarks: boolean; maxDetected: number; minConfidence: number; rotation: boolean; skeleton: { modelPath: string }; skipFrames: number }

Controlls and configures all hand detection specific options

enabled: true/false
landmarks: detect hand landmarks or just hand boundary box
modelPath: paths for hand detector and hand skeleton models, can be absolute path or relative to modelBasePath
minConfidence: threshold for discarding a prediction
iouThreshold: ammount of overlap between two detected objects before one object is removed
maxDetected: maximum number of hands detected in the input, should be set to the minimum number for performance
rotation: use best-guess rotated hand image or just box with rotation as-is, false means higher performance, but incorrect finger mapping if hand is inverted

Type declaration

detector: { modelPath: string }
- modelPath: string
enabled: boolean
iouThreshold: number
landmarks: boolean
maxDetected: number
minConfidence: number
rotation: boolean
skeleton: { modelPath: string }
- modelPath: string
skipFrames: number

modelBasePath

modelBasePath: string

Base model path (typically starting with file://, http:// or https://) for all models

individual modelPath values are relative to this path

object

object: { enabled: boolean; iouThreshold: number; maxDetected: number; minConfidence: number; modelPath: string; skipFrames: number }

Controlls and configures all object detection specific options

enabled: true/false
modelPath: object detection model, can be absolute path or relative to modelBasePath
minConfidence: minimum score that detection must have to return as valid object
iouThreshold: ammount of overlap between two detected objects before one object is removed
maxDetected: maximum number of detections to return

Type declaration

enabled: boolean
iouThreshold: number
maxDetected: number
minConfidence: number
modelPath: string
skipFrames: number

segmentation

segmentation: { enabled: boolean; modelPath: string }

Controlls and configures all body segmentation module removes background from input containing person if segmentation is enabled it will run as preprocessing task before any other model alternatively leave it disabled and use it on-demand using human.segmentation method which can remove background or replace it with user-provided background

enabled: true/false
modelPath: object detection model, can be absolute path or relative to modelBasePath

Type declaration

enabled: boolean
modelPath: string

skipFrame

skipFrame: boolean

Cache sensitivity

values 0..1 where 0.01 means reset cache if input changed more than 1%
set to 0 to disable caching

warmup

warmup: string

What to use for human.warmup()

warmup pre-initializes all models for faster inference but can take significant time on startup
only used for webgl and humangl backends

wasmPath

wasmPath: string

Path to *.wasm files if backend is set to wasm