Find products in an image

Python

from channel3_sdk import Channel3
from channel3_sdk.environment import Channel3Environment

client = Channel3(
    api_key="<value>",
    environment=Channel3Environment.PRODUCTION,
)

client.experimental.find_products_in_image()

const options = {
  method: 'POST',
  headers: {'x-api-key': '<api-key>', 'Content-Type': 'application/json'},
  body: JSON.stringify({})
};

fetch('https://api.trychannel3.com/experimental/find-products-in-image', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

{
  "things": [
    {
      "thing": "<string>",
      "bbox": [
        123
      ],
      "category_ids": [
        "<string>"
      ],
      "gender": "male",
      "matches": [
        {
          "id": "<string>",
          "title": "<string>",
          "description": "<string>",
          "brands": [
            {
              "id": "<string>",
              "name": "<string>"
            }
          ],
          "images": [],
          "category": {
            "slug": "<string>",
            "title": "<string>",
            "has_children": true,
            "path": [
              {
                "slug": "<string>",
                "title": "<string>"
              }
            ]
          },
          "gender": "male",
          "age": "newborn",
          "materials": [
            "<string>"
          ],
          "key_features": [
            "<string>"
          ],
          "colors": [
            {
              "hex": "<string>",
              "percentage": 123
            }
          ],
          "offers": [
            {
              "url": "<string>",
              "domain": "<string>",
              "price": {
                "price": 123,
                "currency": "<string>",
                "compare_at_price": 123
              },
              "availability": "InStock",
              "condition": "new",
              "max_commission_rate": 0,
              "dimensions": {
                "length": {
                  "number": 123,
                  "unit": "mm"
                },
                "width": {
                  "number": 123,
                  "unit": "mm"
                },
                "height": {
                  "number": 123,
                  "unit": "mm"
                },
                "weight": {
                  "number": 123,
                  "unit": "mg"
                }
              }
            }
          ],
          "variants": {
            "options": [
              {
                "name": "<string>",
                "values": [
                  {
                    "label": "<string>",
                    "exists": true,
                    "available": "InStock",
                    "thumbnail_url": "<string>",
                    "product_id": "<string>"
                  }
                ]
              }
            ],
            "selected": [
              {
                "name": "<string>",
                "label": "<string>"
              }
            ]
          },
          "structured_attributes": {},
          "ratings": {
            "value": 123,
            "count": 123,
            "source_count": 123
          }
        }
      ],
      "masks": [
        {
          "bbox": [
            123
          ],
          "rle": {
            "size": [
              123
            ],
            "counts": "<string>"
          }
        }
      ]
    }
  ],
  "image_size": [
    123
  ]
}

POST

experimental

find-products-in-image

Python

from channel3_sdk import Channel3
from channel3_sdk.environment import Channel3Environment

client = Channel3(
    api_key="<value>",
    environment=Channel3Environment.PRODUCTION,
)

client.experimental.find_products_in_image()

const options = {
  method: 'POST',
  headers: {'x-api-key': '<api-key>', 'Content-Type': 'application/json'},
  body: JSON.stringify({})
};

fetch('https://api.trychannel3.com/experimental/find-products-in-image', options)
  .then(res => res.json())
  .then(res => console.log(res))
  .catch(err => console.error(err));

{
  "things": [
    {
      "thing": "<string>",
      "bbox": [
        123
      ],
      "category_ids": [
        "<string>"
      ],
      "gender": "male",
      "matches": [
        {
          "id": "<string>",
          "title": "<string>",
          "description": "<string>",
          "brands": [
            {
              "id": "<string>",
              "name": "<string>"
            }
          ],
          "images": [],
          "category": {
            "slug": "<string>",
            "title": "<string>",
            "has_children": true,
            "path": [
              {
                "slug": "<string>",
                "title": "<string>"
              }
            ]
          },
          "gender": "male",
          "age": "newborn",
          "materials": [
            "<string>"
          ],
          "key_features": [
            "<string>"
          ],
          "colors": [
            {
              "hex": "<string>",
              "percentage": 123
            }
          ],
          "offers": [
            {
              "url": "<string>",
              "domain": "<string>",
              "price": {
                "price": 123,
                "currency": "<string>",
                "compare_at_price": 123
              },
              "availability": "InStock",
              "condition": "new",
              "max_commission_rate": 0,
              "dimensions": {
                "length": {
                  "number": 123,
                  "unit": "mm"
                },
                "width": {
                  "number": 123,
                  "unit": "mm"
                },
                "height": {
                  "number": 123,
                  "unit": "mm"
                },
                "weight": {
                  "number": 123,
                  "unit": "mg"
                }
              }
            }
          ],
          "variants": {
            "options": [
              {
                "name": "<string>",
                "values": [
                  {
                    "label": "<string>",
                    "exists": true,
                    "available": "InStock",
                    "thumbnail_url": "<string>",
                    "product_id": "<string>"
                  }
                ]
              }
            ],
            "selected": [
              {
                "name": "<string>",
                "label": "<string>"
              }
            ]
          },
          "structured_attributes": {},
          "ratings": {
            "value": 123,
            "count": 123,
            "source_count": 123
          }
        }
      ],
      "masks": [
        {
          "bbox": [
            123
          ],
          "rle": {
            "size": [
              123
            ],
            "counts": "<string>"
          }
        }
      ]
    }
  ],
  "image_size": [
    123
  ]
}

Pass config.collection_id to search each detected thing inside a collection.

Authorizations

x-api-key

string

header

required

Headers

x-user-id

string | null

Optional user identifier to attribute clicks and sales to a user in your system. Channel3 appends it to buy URLs in the response.

Body

application/json

Find purchasable products in a scene image, grouped by detected thing.

image_url

string | null

Publicly accessible URL of the image.

base64_image

string | null

Base64 encoded image bytes (no data URI prefix).

limit

integer

default:10

Max catalog matches per detected thing.

Required range: 1 <= x <= 30

filters

SearchFilters · object

Applied identically to every per-thing catalog search.

Show child attributes

config

BaseConfig · object

Locale and optional collection scope.

Show child attributes

max_things

integer

default:6

Cap on named objects to detect and search.

Required range: 1 <= x <= 6

return_masks

boolean

default:false

Include each thing's segmentation mask in masks. masks is null when the detector produced no mask for a thing.

Response

Successful Response

things

DetectedThing · object[]

required

Show child attributes

image_size

integer[]

required

[width, height] of the image the boxes are relative to.

Products

Conversational Agent

Collections

Brands

Categories

Websites

Reporting

Price tracking

Models

Find products in an image

Authorizations

Headers

Body

Response