NvidiaPatent applicationAI & softwareInterest 6/10

Nvidia patents method to merge visual features for autonomous navigation

Patent drawing
Patent drawing from the official filing.

Nvidia describes a system that combines two different methods for spotting details in images to help machines navigate. It uses one model to follow specific points from one image to the next and another to find new points, then merges these to create a more reliable map of the environment. The system can prioritize certain areas, such as the road, over others like the sky, to improve tracking accuracy.

Why it mattersThis could make self-driving cars and robots better at understanding their surroundings by filtering out less important visual details.

In the filing’s own words

In various examples, feature tracking for autonomous or semi-autonomous systems and applications is described herein. Systems and methods are disclosed that merge, using one or more processes, features detected using a feature tracker(s) and features detected using a feature detector(s) in order to track features between images. In some examples, the number of merged features and/or the locations of the merged features within the images are limited. This way, the systems and methods are able to identify merged features that are of greater importance for tracking while refraining from tracking merged features that are of less importance. For example, if the systems and methods are being used to identify features for autonomous driving, a greater number of merged features that are associated with objects located proximate to the driving surface may be tracked as compared to merged features that are associated with the sky.

What exactly they claim (claim 1)

1. A method comprising: tracking, using a first machine learning model, a tracked feature point between a first image and a second image; detecting, using a second machine learning model, a detected feature point in the second image that is separate from the tracked feature point; determining at least a merged feature point within the second image that represents the tracked feature point and the detected feature point; and causing, based at least on the merged feature point, a machine to navigate within an environment.

Facts

Company
Nvidia (NVIDIA Corporation)
Official title
MERGING TRACKED FEATURES
Publication number
US20260360116A1
Application number
19747841
Filed
Jul 20, 2026
Published
Nov 12, 2026
Inventors
4
Claims
20

Classified under

G06V 10/771 Physics & computingG06V 10/7715 Physics & computing

More from Nvidia in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.