US 11,900,611 B2 · Generating object masks of object parts utlizing deep learning

Abstract

The present disclosure relates to a class-agnostic object segmentation system that automatically detects, segments, and selects objects within digital images irrespective of object semantic classifications. For example, the object segmentation system utilizes a class-agnostic object segmentation neural network to segment each pixel in a digital image into an object mask. Further, in response to detecting a selection request of a target object, the object segmentation system utilizes a corresponding object mask to automatically select the target object within the digital image. In some implementations, the object segmentation system utilizes a class-agnostic object segmentation neural network to detect and automatically select a partial object in the digital image in response to a target object selection request.

Claims20

Dependent claims

A computer-implemented method comprising:

displaying a digital image comprising a plurality of objects;
generating object masks for the plurality of objects utilizing an object segmentation neural network;
detecting a first selection request on an object in the digital image;
based on detecting the first selection request, providing an object mask of the object from the object masks generated for the digital image;
detecting a second selection request on the object;
and
in response to the second selection request, providing a second object mask of a second object that forms part of the object.

4 more elements

Dependencies are inferred from the claim text.

Generating object masks of object parts utlizing deep learning

Life of this patent

Generating object masks of object parts utlizing deep learning

Life of this patent

Abstract

Claims20