CarMax. has been granted a patent for a system that generates synthetic images using a method involving three neural networks. The system trains these networks to create images that meet specific classification requirements, ensuring high accuracy in matching real images and transforming latent feature vectors. GlobalData’s report on CarMax gives a 360-degree view of the company including its patenting strategy. Buy the report here.

Smarter leaders trust GlobalData

Report-cover

Data Insights CarMax Inc - Company Profile

Buy the Report

Data Insights

The gold standard of business intelligence.

Find out more

According to GlobalData’s company profile on CarMax, was a key innovation area identified from patents. CarMax's grant share as of June 2024 was 92%. Grant share is based on the ratio of number of grants to total number of patents.

System for generating synthetic images using neural networks

Source: United States Patent and Trademark Office (USPTO). Credit: CarMax Inc

The patent US12039647B1 describes a sophisticated system and method for generating synthetic images using multiple neural networks. The system comprises three interconnected neural networks: the first generates synthetic images that closely resemble real images, the second distinguishes between synthetic and real images, and the third transforms synthetic images into latent feature vectors. The process involves training these networks simultaneously, allowing the first network to create a synthetic image based on a latent feature vector, while the third network refines this vector by incorporating target features from a provided target image. The system can manipulate various features of the synthetic images, such as color, orientation, lighting, and background, to produce a second synthetic image that includes specific target features.

Additionally, the patent outlines a method for generating synthetic images that includes receiving target images with labeled features, converting these images into latent feature vectors, and transforming a first image (which lacks the labeled feature) into a new latent feature vector through logistic regression. This transformation enables the generation of a synthetic image that incorporates the desired labeled feature. The method emphasizes the ability to manipulate features of the latent vectors to achieve the desired characteristics in the synthetic images, particularly in applications involving vehicles. Overall, the patent presents a comprehensive framework for enhancing synthetic image generation through advanced neural network training and feature manipulation techniques.

To know more about GlobalData’s detailed insights on CarMax, buy the report here.

Data Insights

From

The gold standard of business intelligence.

Blending expert knowledge with cutting-edge technology, GlobalData’s unrivalled proprietary data will enable you to decode what’s happening in your market. You can make better informed decisions and gain a future-proof advantage over your competitors.

GlobalData

GlobalData, the leading provider of industry intelligence, provided the underlying data, research, and analysis used to produce this article.

GlobalData Patent Analytics tracks bibliographic data, legal events data, point in time patent ownerships, and backward and forward citations from global patenting offices. Textual analysis and official patent classifications are used to group patents into key thematic areas and link them to specific companies across the world’s largest industries.