Facebook, Inc.

Michael Oren Belkin

Christopher Joseph Welch

An online system is configured to provide content items to users. The content item includes an image, and is displayed with an interface element colored using an accent color to create a unified look and feel with the displayed image. The accent color is dynamically selected based upon the image, extracted color features of the image, and embeddings associated with the image indicating at least one object depicted in the image. A machine-trained classification model selects the color to be used in displaying the interface element from a quantized set of colors of the image, based upon the extracted color features and the embeddings associated with the image. As such, suitable accent colors can be selected automatically for large numbers of content items, in a flexible manner that can account for the context of the images and the context in which the content item is to be displayed.

Color sampling selection for displaying content items using machine learning

An online system is configured to provide content items to users. The content item includes an image, and is displayed with a color block or colored border as an accent to the image. The color of the color block or colored bordered is dynamically selected based upon the image included in the content item, and is selected using a waterfall selection method comprising a sequence of different selection methods, each selection method configured to select a color based upon a most common color of the image that satisfies a respective set of selection criteria or determine that no color meets the respective set of selection criteria. The selection methods may comprise a dominant color selection method wherein the selected color corresponds to a most dominant color of the image, and a complementary color selection method wherein the select color is a complement of a dominant color of the image.

Growing community of inventors

San Francisco, CA, United States of America

Michael Oren Belkin