Microsoft technology licensing, llc (20240119695). GENERATION OF EMPHASIS IMAGE WITH EMPHASIS BOUNDARY simplified abstract

GENERATION OF EMPHASIS IMAGE WITH EMPHASIS BOUNDARY

Organization Name

microsoft technology licensing, llc

Inventor(s)

Salman Muin Kayser Chishti of Tallinn (EE)

GENERATION OF EMPHASIS IMAGE WITH EMPHASIS BOUNDARY - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240119695 titled 'GENERATION OF EMPHASIS IMAGE WITH EMPHASIS BOUNDARY

Simplified Explanation

The automated generation of an emphasis image based on an input image involves feeding the input image to a machine-learned model trained to label portions of images. The model outputs identifications and labels for these portions, which are used to determine an emphasis bounding box applied to the input image to generate an emphasis image, such as a cropped image.

• Machine-learned model trained to label portions of images
• Output includes identifications and labels for identified portions
• Emphasis bounding box determined based on generated labels
• Emphasis bounding box applied to input image to create emphasis image

Potential Applications

This technology could be applied in image editing software to automatically crop images based on the identified portions, making it easier for users to highlight specific elements in their photos.

Problems Solved

This technology solves the problem of manually cropping images to emphasize certain elements, saving time and effort for users who want to quickly create impactful visuals.

Benefits

The benefits of this technology include increased efficiency in image editing tasks, as well as the ability to easily highlight important aspects of an image without the need for manual intervention.

Potential Commercial Applications

A potential commercial application of this technology could be in social media platforms or online marketplaces, where users often need to quickly enhance and highlight images for better engagement or sales.

Possible Prior Art

One possible prior art for this technology could be existing image editing software that offers automated cropping features based on predefined rules or user input.

How does the machine-learned model determine the labels for the identified portions of the image?

The process of labeling the identified portions by the machine-learned model is not detailed in the abstract. It would be interesting to know more about the specific algorithms or techniques used for this task.

Are there any limitations to the size or complexity of images that can be processed using this technology?

The abstract does not mention any restrictions on the size or complexity of images that can be handled by this technology. Understanding the scalability and performance of the system would be crucial for potential users.

Original Abstract Submitted

the automated generation of an emphasis image (such as a cropped image) that is based on an input image. the input image is fed to a machine-learned model that is trained to label portions of images. that machine-learned model then outputs an identification of multiple portions of images, along with potentially labels of each of those identified portions. the label identifies a property of the corresponding identified portion. as an example, one portion might be labelled as irrelevant, another might be labelled as a name, another might be labelled as a comment, and so forth. that output is accessed and the generated label is used to determine an emphasis bounding box. the emphasis bounding box is then applied to the input image to generate an emphasis image. as an example, the emphasis image may be a cropped image of the input image.