Thread Genius recently launched with a goal of improving content recommendation based on computer vision. At the heart of their strategy is the Thread Genius API. The Thread Genius API drives what the company has termed visual style recommendation as a service. The service helps API users make product recommendations to consumers based on the consumers' taste through the use of images instead of words.
"The crux behind recommendations, whether by a human or machine, is the ability to articulate taste," the company explains. "Thread Genius was founded on this principle that no amount of words can fully articulate our taste. This is why when we're asked about our taste in music, we play songs. For fashion, we show pictures."
The company was born this year out of Techstars '17. Thread Genius believes its computer vision artificial intelligence can improve the recommendations that drive much of ecommerce business today. At the broadest level, Thread Genius hopes to "map the world's visual taste."
After integrating the Thread Genius API, app and service users can tag photos with certain stylistic elements, perform visual search, and other computer vision based operations that prevent services from having to generate recommendations based on word terms alone. The API is RESTful and returns calls in a JSON data format.
Within the API, developers can input images locally or from a URL. Once images are uploaded, cropping functions are also available. For those looking to expand use of Thread Genius, the Catalog API allows integrating parties to create searchable catalogs of objects. Then, the Prediction API allows developers to provide tags and bounding boxes around specific items. For more details, check out the API docs.