🟡API for Masks

This page contains description how to get images with applied masks via API.

Masks are additional results of image analysis, representing the spatial information about detected skin/hair health returned by the selected set of Algorithms.

Check the Masks description for more details.

What can you do?

You can find more details on these actions in the Rasterisation section of Swagger.

• Apply masks to image with API

---

Apply masks to image with API

Apply Mask V4 Http

Possible algorithms: Algorithms with available concentric_lines option (Note: concentric_lines are available only for restored image_type):

• acne
• translucency
• pigmentation
• redness
• hydration

Algorithms with available vector masks (including previous list):

• pores
• uniformness
• eye_bags
• lines

Algorithms that will return only metrics (including previous lists):

• age (only main_metric)
• skin_tone

You can request metrics for each algorithm by setting main_metric and sub_metrics to true.

POST/service/rasterisation/v4/masks/

Authorization

HTTPBearer

cookie

Query parameters

Body

application/json

image_id*Image Id

image_type*ImageType

originalrestored

masks*Masks

Response

200422

Successful Response

Body

application/json

algorithm_tech_name*Algorithm Tech Name

masked_image_urlMasked Image Url

stringnull

errorError

stringnull

main_metricany of

SubMetricV3null

sub_metricsSub Metrics

feedbackFeedback

Request

JavaScriptCurlPython

const response = await fetch('/service/rasterisation/v4/masks/?company_id=123e4567-e89b-12d3-a456-426614174000', {
    method: 'POST',
    headers: {
      "Authorization": "Bearer <token>",
      "Content-Type": "application/json"
    },
    body: JSON.stringify({
      "image_id": "text",
      "image_type": "original",
      "masks": [
        {
          "algorithm_tech_name": "text"
        }
      ]
    }),
});
const data = await response.json();

Response

200422

[
  {
    "algorithm_tech_name": "text",
    "masked_image_url": "text",
    "error": "text",
    "main_metric": {
      "name": "text",
      "tech_name": "text",
      "widget_type": "text"
    },
    "sub_metrics": [
      {
        "name": "text",
        "tech_name": "text",
        "widget_type": "text"
      }
    ]
  }
]

Required arguments are company_id and image_id.

It's important which image you want to visualize masks on: original or restored.

Then you need to compile a masks list:

• algorithm_tech_name - it's internal ML algorithm tech_name (e.g. age or pores). We know that this is an incosistency, because we have 2 types of tech names throughout API. It will be changed in future.

• optional fill_override - override mask color.

• optional polyline_line_thickness - override mask line thickness.

• optional main_metric - if API response should contain the metric for this algorithm.

• optional sub_metrics - if API response should contain submetrics for this algorithm.

• optional deprecated concentric_lines - if masks should be visualised as concentric lines. (Supported only for a few algorithms). Will be deleted in future.

This method returns signed URLs with images with applied masks.

Note We return 1 image per 1 mask, so if you request to apply 5 masks, you will receive 5 images, not 1 in response.