This endpoint retrieves all of the given attributes (from requested_attributes) that can be extracted from the provided image and text. If no attributes are given to the endpoint, then the API will attempt to extract whatever is possible from the inputs and return their associated values. Either the product name, image url, or product description must be provided to the API.
Attributes are extracted from an image and text by employing machine learning models that rely on named entity recognition and image classification techniques. To ensure high quality, we require that only valid attributes are provided as input to this endpoint. Any other provided attributes will be ignored and will not be retrieved. You can find definitions and examples of the valid attributes here. Additionally, only English text is supported as text inputs to this endpoint, as the model will not perform well on non-English text.
See the Attributes Sample for a sample of available attributes.