Pores Result Scheme

Haut.AI provides an AI algorithm that estimates pores in different facial areas and the whole face.

Description

Pores Algorithm

algorithm_tech_name

pores

area_name

The algorithm returns metrics and sub-metrics for the following facial areas:

• face

• forehead

• nose

• right_cheek

• left_cheek

main_metric

The main metric is an overall score indicating the pores for each area_name

"main_metric": {
                "value": 64,
                "widget_type": "bad_good_line",
                "name": "Pores Score",
                "tech_name": "Pores_score",
                "widget_meta": null
            },

• value the pores score ranges from [0,100] More large pores are associated with a lower value for this parameter

• "widget_type": "bad_good_line" indicates that a higher value corresponds to a better skin condition. This score can be classified into six classes based on value:

  • (90,100] - Excellent

  • (80,90] - Great

  • (70,80] - Good

  • (50,70] - Average

  • (30, 50] - Poor

  • [0,30] - Bad

• "name": "Pores score"

• "tech_name": "pores_score"

sub_metrics

The algorithm returns sub-metrics for "area_name": face only

"sub_metrics": [
                {
                    "value": 455,
                    "widget_type": "numeric",
                    "name": "Pores Number",
                    "tech_name": "pores_number",
                    "widget_meta": null
                },
                {
                    "value": 1,
                    "widget_type": "density",
                    "name": "Pores Density",
                    "tech_name": "pores_density",
                    "widget_meta": null
                }

• Pores Number estimates the number of detected pores

  • value The number of detected pores

  • "widget_type": "numeric"

  • "name": "Pores Number"

  • "tech_name": "pores_number"

• Pores Density estimates the density of detected pores

  • value the density of detected pores on the skin

  • "widget_type": "density"

  • "name": "Pores Density"

  • "tech_name": "pores_density"

masks_restored

The algorithm returns a vectorized mask of pores for a face-aligned image

• "mask_type": "point_mask"

• features

  • geometry

    • "type": "MultiPoint"

  See more on Masks here

masks_original

The algorithm returns a vectorized mask of pores for an original image

• "mask_type": "point_mask"

• features

  • geometry

    • "type": "MultiPoint"

  See more on Masks here

Example (JSON)

431KB

pores.json

pores.json