Generating object-based layers for digital image editing using object classification machine learning models

Abstract

The present disclosure relates to systems, methods, and non-transitory computer readable media for accurately, efficiently, and flexibly generating image layers and determining layer labels utilizing a machine learning approach. For example, the disclosed systems utilize an image segmentation machine learning model to segment the digital image and identify individual objects depicted within the digital image. Additionally, in some embodiments, the disclosed systems determine object classifications for the depicted objects by utilizing an object classification machine learning model. In some cases, the disclosed systems further generate image layers for the digital image by generating a separate layer for each identified object (or for groups of similar objects). In certain embodiments, the disclosed systems also determine layer labels for the image layers according to the object classifications of the respective objects depicted in each of the image layers.