Granted Patent B2

US 12,560,817 · App. 19/072,793 · Granted Feb 24, 2026

Inventors: Jouya Jadidian (Los Gatos, CA); Calin Cristian (Iasi, RO); Seyedsohrab Madani (Menlo Park, CA); Mohit Narang (Cupertino, CA)

Assignee: Rivet Industries, Inc.

G02B27/0172G06T5/70G06T7/143G06V20/20G06V20/41G06V20/64G02B2027/0112G02B2027/0138G02B2027/014G02B2027/0178G06T2207/10024G06T2207/30196

Abstract

Claims (68)

1 . A head-mounted display (HMD), comprising:

a processor;

a display operatively coupled to the processor;

a camera operatively coupled to the processor;

an inertial measurement unit (IMU) operatively coupled to the processor; and

a memory operatively coupled to the processor, the processor configured to:

receive a plurality of images captured by the camera under a low light condition;

apply at least one of a denoising technique or a filtering technique to the plurality of images to generate a plurality of modified images;

identify a plurality of objects included in at least one modified image from the plurality of modified images, at least one object from the plurality of objects included in at least two modified images from the plurality of modified images;

for each object from the plurality of objects and to generate a plurality of color images, assign a color from a plurality of colors to that object for each modified image from the plurality of modified images including that object, each object from the plurality of objects assigned to a color from the plurality of colors that is different from each remaining color from the plurality of colors;

cause the plurality of color images to be displayed via the display;

receive, via the IMU, pose data associated with a modified orientation of the HMD;

modify the plurality of color images based on the pose data to generate an updated plurality of color images that maintain color consistency of objects from the plurality of objects across the updated plurality of color images; and

display the updated plurality of color images via the display.

2 . The HMD of claim 1 , wherein the HMD is a three-degree of freedom (3DOF) virtual reality headset.

3 . The HMD of claim 1 , wherein the HMD is a six-degree of freedom (6DOF) virtual reality headset.

4 . The HMD of claim 1 , wherein the plurality of images are black-and-white images.

5 . The HMD of claim 1 , wherein an illuminance of the low light condition is not more than at least one of one lumen or one lux.

6 . The HMD of claim 1 , wherein the plurality of images are associated with a first brightness and the plurality of modified images are associated with a second brightness higher than the first brightness.

7 . The HMD of claim 1 , wherein the plurality of images are associated with a first contrast and the plurality of modified images are associated with a second contrast greater than the first contrast.

8 . The HMD of claim 1 , wherein the processor is further configured to:

determine a scene associated with the plurality of images; and

determine the at least one of the denoising technique or the filtering technique based on the scene.

9 . The HMD of claim 1 , wherein the plurality of images are associated with a scene and the processor is further configured to:

receive user input indicating a set of attributes associated with the scene, the set of attributes including at least one of a light level or an environment type; and

determine, before applying the at least one of the denoising technique or the filtering technique, the at least one of the denoising technique or the filtering technique based on the set of attributes associated with the scene.

10 . The HMD of claim 1 , wherein the processor is configured to, when identifying the plurality of objects included in the plurality of modified images:

apply semantic segmentation to the plurality of modified images.

11 . The HMD of claim 1 , wherein the processor is configured to, when identifying the plurality of objects included in the plurality of modified images:

apply panoptic segmentation to the plurality of modified images.

12 . A method, comprising:

receiving a first image of a first scene captured via a head-mounted display (HMD) under a low light condition;

applying at least one of a first denoising technique or a first filtering technique to the first image to generate a first modified image;

identifying a first object in the first modified image, the first modified image not including a second object;

in response to determining that the first object does not have a color assignment, assigning a first color and not a second color to the first object at the first modified image;

receiving a second image of the first scene captured via the HMD under the low light condition;

applying the at least one of the first denoising technique or the first filtering technique to the second image to generate a second modified image;

identifying (1) the first object in the second modified image and (2) the second object in the second modified image;

in response to determining that the second object does not have the color assignment, assigning the second color and not the first color to the second object at the second modified image;

in response to determining that the first object is assigned the first color at the first modified image, assigning the first color and not the second color to the first object at the second modified image;

receiving a third image captured via the HMD under the low light condition, the third image of a second scene different than the first scene;

applying at least one of a second denoising technique or a second filtering technique to the third image to generate a third modified image;

identifying a third object in the third modified image; and

in response to determining that the third object does not have the color assignment, assigning a third color to the third object at the third modified image.

13 . The method of claim 12 , wherein the first object is at a first location in the first modified image and the first object is at a second location in the second modified image different than the first location.

14 . The method of claim 12 , wherein the first image is captured when the HMD is at a first orientation and the second image is captured when the HMD is at a second orientation different than the first orientation.

15 . The method of claim 12 , further comprising:

receiving a fourth image captured via the HMD under the low light condition;

applying the at least one of the denoising technique or the filtering technique to the fourth image to generate a fourth modified image;

identifying (1) the first object in the fourth modified image and (2) the second object in the fourth modified image and (3) a fourth object in the fourth modified image;

in response to determining that the fourth object does not have the color assignment, assigning a fourth color and not the first color or the second color to the fourth object at the fourth modified image;

in response to determining that the first object is assigned the first color at least one of the first modified image or the second modified image, assigning the first color and not the second color or the fourth color to the first object at the fourth modified image; and

in response to determining that the second object is assigned the second color at the second modified image, assigning the second color and not the first color or the fourth color to the second object at the fourth modified image.

16 . The method of claim 12 , further comprising:

receiving a fourth image captured via the HMD under the low light condition, the fourth image not including the first object and the second object;

applying the at least one of the denoising technique or the filtering technique to the fourth image to generate a fourth modified image;

identifying a fourth object in the fourth modified image; and

in response to determining that the fourth object does not have the color assignment, assigning a fourth color and not the first color or the second color to the fourth object at the fourth modified image.

17 . The method of claim 12 , wherein the first object has an actual color at a scene that includes the first object, and the first color assigned to the first object is not the actual color.

18 . A non-transitory processor-readable medium storing code representing instructions to be executed by one or more processors, the instructions comprising code to cause the one or more processors to:

receive a plurality of black-and-white images captured without active illumination;

apply at least one of a denoising technique or a filtering technique to the plurality of black-and-white images to generate a plurality of modified images;

identify a plurality of objects included in the plurality of modified images, at least one object from the plurality of objects included in at least two modified images from the plurality of modified images;

for each object from the plurality of objects and to generate a plurality of color images, assign a color from a plurality of colors to that object for each modified image from the plurality of modified images including that object, each object from the plurality of objects assigned to a color different from the remaining colors from the plurality of colors;

cause the plurality of color images to be displayed at a display;

receive pose data associated with a modified orientation of a head-mounted display (HMD); and

modify the plurality of color images based on the pose data to generate an updated plurality of color images that maintain color consistency of objects from the plurality of objects across the updated plurality of color images.

19 . The non-transitory processor-readable medium of claim 18 , wherein (1) the plurality of black-and-white images are associated with a first brightness and the plurality of modified images are associated with a second brightness greater than the first brightness and (2) the plurality of black-and-white images are associated with a first contrast and the plurality of modified images are associated with a second contrast greater than the first contrast.

Assignments (3)

Continuity (2)

References Cited (35)