Object oriented image editing

A non-transitory computer-readable medium storing instructions that, when executed by one or more processors, cause the one or more processors to perform operations comprising:

• segmenting a first image into regions depicting objects;
• identifying two or more different classifications descriptive of respective objects of the objects in the first image;
• identifying at least one subcategory for a respective classification of the two or more different classifications, wherein the at least one subcategory has a greater level of specificity than a specificity of the respective classification and the at least one subcategory is represented by a respective semantic term;
• storing locator information for regions of the firstimage corresponding to the respective objects associated with the at least one subcategory;
• providing to a user device, the respective semantic term of the at least one sub category;
• receiving a user selection of at least one semantic term indicating a select subcategory of the at least one subcategory;
• receiving a user request to apply one or more edit operations to at least one instance of an object in the firstimage associated with the select subcategory;
• and
• applying the one or more edit operations to the at least one instance of the object in the firstimage associated with the select subcategory by locating the at least one instance of the object with the location information for the at least one instance of the object.

A system comprising:

• one or more processors;
• and
• one or more computer-readable media having instructions stored thereon that, when executed by the one or more processors, cause performance of operations comprising:
• segmenting a first image into regions depicting objects;
• identifying two or more different classifications descriptive of respective objects of the objects in the first image;
• identifying at least one subcategory for a respective classification of the two or more different classifications, wherein the at least one subcategory has a greater level of specificity than a specificity of the respective classification and the at least one subcategory is represented by a respective semantic term;
• storing locator information for regions of the first image corresponding to the respective objects associated with the at least one subcategory;
• providing to a user device, the respective semantic term of the at least one sub category;
• receiving a user selection of at least one semantic term indicating a select subcategory of the at least one subcategory;
• receiving a user request to apply one or more edit operations to at least one instance of an object in the first image associated with the select subcategory;
• and
• applying the one or more edit operations to the at least one instance of the object in the first image associated with the select subcategory by locating the at least one instance of the object with the location information for the at least one instance of the object.