US 2026/0228910 A1Published application

Multi-level local feature detection and extraction

AssigneeQUALCOMM Incorporated

InventorsIago SUAREZ, Andrii MAIDAN, Kostiantyn ANTONIUK, Maksym RAITAROVSKYI

Life of this patent

Estimating the term…

Loading patent…

Abstract

The present disclosure provides techniques for feature detection and extraction. An example method includes obtaining a downsampled image comprising first pixels at a first resolution that is less than a second resolution; generating, with a neural network, for each pixel of the first pixels, a first respective probability that the pixel is a key point; upsampling the downsampled image to a second image comprising second pixels at a third resolution that is greater than the first resolution, wherein each pixel of the second pixels is associated with a corresponding pixel of the first pixels and the first respective probability associated with the corresponding pixel; generating, with the neural network, for each pixel of the second pixels for which the associated first respective probability is greater than a threshold probability, a second respective probability that the pixel is a key point; extracting a respective descriptor; and executing a localization task.

Claims20

Dependent claims

An apparatus, comprising:

a processing system that includes one or more processors and one or more memories coupled with the one or more processors, the processing system configured to cause the apparatus to:
obtain a downsampled image of an image of an environment, wherein the downsampled image comprises a plurality of first pixels at a first resolution that is less than a second resolution of the image;
generate, with a neural network, for each pixel of the plurality of first pixels, a first respective probability that the pixel is a key point;
upsample the downsampled image to a second image comprising a plurality of second pixels at a third resolution that is greater than the first resolution, wherein each pixel of the plurality of second pixels is associated with a corresponding pixel of the plurality of first pixels and the first respective probability associated with the corresponding pixel;
generate, with the neural network, for each pixel of the plurality of second pixels for which the associated first respective probability is greater than a threshold probability, a second respective probability that the pixel is a key point;
extract, for one or more second pixels of the plurality of second pixels for which the associated second respective probability is greater than the threshold probability, a respective descriptor;
and
execute a localization task configured to utilize the respective descriptor of each of the one or more second pixels.

5 more elements

A method for feature detection and extraction by an apparatus, comprising:

obtaining a downsampled image of an image of an environment, wherein the downsampled image comprises a plurality of first pixels at a first resolution that is less than a second resolution of the image;
generating, with a neural network, for each pixel of the plurality of first pixels, a first respective probability that the pixel is a key point;
upsampling the downsampled image to a second image comprising a plurality of second pixels at a third resolution that is greater than the first resolution, wherein each pixel of the plurality of second pixels is associated with a corresponding pixel of the plurality of first pixels and the first respective probability associated with the corresponding pixel;
generating, with the neural network, for each pixel of the plurality of second pixels for which the associated first respective probability is greater than a threshold probability, a second respective probability that the pixel is a key point;
extracting, for one or more second pixels of the plurality of second pixels for which the associated second respective probability is greater than the threshold probability, a respective descriptor;
and
executing a localization task configured to utilize the respective descriptor of each of the one or more second pixels.

4 more elements

Dependencies are inferred from the claim text.