IP Library Granted Patent US 12,475,583
Granted Patent B2
US 12,475,583 · App. 18/309,603 · Granted Nov 18, 2025

Stereo vision with census transform

Inventors: James Wilson Nash (San Diego, CA); Kalin Mitkov Atanassov (San Diego, CA); Jason Chung (San Diego, CA)
Assignee: QUALCOMM Incorporated
G06T7/521G06T7/593G06V10/141G06V10/25G06T2207/10028
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,475,583
App. No.
18/309,603
Granted
Nov 18, 2025
Kind
B2
Abstract

Systems and techniques are provided for wireless communication. For example, a process can include obtaining a first image of an environment, the first image including a pattern of light. The process can further include determining a pixel window for a pixel of the first image; determining a set of transform pixels based on the pattern of light, wherein the set of transform pixels are a subset of pixels from the pixel window; performing a census transform based on the pixel and the set of transform pixels to obtain census transform information for the first image; and generating depth data based on the census transform information for the first image and census transform information for a second image.

Claims (48)

1 . A device for image processing, comprising:

at least one memory;

a light source configured to project a pattern of light into an environment; and

at least one processor coupled to the at least one memory, the at least one processor configured to:

obtain a first image of the environment, the first image including the pattern of light, wherein the first image is captured by a camera;

determine a pixel window for a pixel of the first image, the pixel window comprising a set of pixels around the pixel;

determine a set of transform pixels based on the pattern of light projected by the light source, wherein the set of transform pixels are a subset of pixels from the pixel window;

perform a census transform based on the pixel and the set of transform pixels to obtain census transform information for the first image; and

generate depth data based on the census transform information for the first image.

2 . The device of claim 1 , wherein the pattern of light is in a polygonal pattern.

3 . The device of claim 2 , wherein dots of light of the pattern of light appear in portions of the polygonal pattern for the pixel window.

4 . The device of claim 3 , wherein the dots of light appear at vertices of the polygonal pattern.

5 . The device of claim 4 , wherein the set of transform pixels correspond to the vertices of the polygonal pattern.

6 . The device of claim 4 , wherein the dots of light appear at a subset of all of the vertices of the polygonal pattern.

7 . The device of claim 2 , wherein the polygonal pattern comprises a hexagonal pattern.

8 . The device of claim 1 , wherein an output of the census transform comprises a vector.

9 . The device of claim 8 , wherein a structure of the vector is based on the pattern of light.

10 . The device of claim 1 , wherein the device comprises a mobile device.

11 . The device of claim 1 , wherein the depth data comprises a depth map.

12 . A method for image processing, comprising:

obtaining a first image of an environment, the first image including a pattern of light projected by a light source, wherein the first image is captured by a camera;

determining a pixel window for a pixel of the first image, the pixel window comprising a set of pixels around the pixel;

determining a set of transform pixels based on the pattern of light projected by the light source, wherein the set of transform pixels are a subset of pixels from the pixel window;

performing a census transform based on the pixel and the set of transform pixels to obtain census transform information for the first image; and

generating depth data based on the census transform information for the first image.

13 . The method of claim 12 , wherein the pattern of light is in a polygonal pattern.

14 . The method of claim 13 , wherein dots of light of the pattern of light appear in portions of the polygonal pattern for the pixel window.

15 . The method of claim 14 , wherein the dots of light appear at vertices of the polygonal pattern.

16 . The method of claim 15 , wherein the set of transform pixels correspond to the vertices of the polygonal pattern.

17 . The method of claim 15 , wherein the dots of light appear at a subset of all of the vertices of the polygonal pattern.

18 . The method of claim 13 , wherein the polygonal pattern comprises a hexagonal pattern.

19 . The method of claim 12 , wherein an output of the census transform comprises a vector.

20 . The method of claim 19 , wherein a structure of the vector is based on the pattern of light.

21 . The method of claim 12 , wherein the depth data comprises a depth map.

22 . A non-transitory computer-readable medium having stored thereon instructions that, when executed by a processor, cause the processor to:

obtain a first image of an environment, the first image including a pattern of light projected by a light source, wherein the first image is captured by a camera;

determine a pixel window for a pixel of the first image, the pixel window comprising a set of pixels around the pixel;

determine a set of transform pixels based on the pattern of light projected by the light source, wherein the set of transform pixels are a subset of pixels from the pixel window;

perform a census transform based on the pixel and the set of transform pixels to obtain census transform information for the first image; and

generate depth data based on the census transform information for the first image.

23 . The non-transitory computer-readable medium of claim 22 , wherein the pattern of light is in a polygonal pattern.

24 . The non-transitory computer-readable medium of claim 23 , wherein dots of light of the pattern of light appear in portions of the polygonal pattern for the pixel window.

25 . The non-transitory computer-readable medium of claim 24 , wherein the dots of light appear at vertices of the polygonal pattern.

26 . The non-transitory computer-readable medium of claim 25 , wherein the set of transform pixels correspond to the vertices of the polygonal pattern.

27 . The non-transitory computer-readable medium of claim 25 , wherein the dots of light appear at a subset of all of the vertices of the polygonal pattern.

28 . The non-transitory computer-readable medium of claim 23 , wherein the polygonal pattern comprises a hexagonal pattern.

29 . The non-transitory computer-readable medium of claim 22 , wherein an output of the census transform comprises a vector.

30 . The non-transitory computer-readable medium of claim 29 , wherein a structure of the vector is based on the pattern of light.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Jun 13, 2023
From: NASH, JAMES WILSON; ATANASSOV, KALIN MITKOV; CHUNG, JASON
To: QUALCOMM INCORPORATED
Reel/Frame 063938/0934 →
Continuity (1)
Related Publication 20240362805A1 · Oct 31, 2024
References Cited (13)
US 10091419B2 · John Archibald · 2018 [cited by examiner]
US 10349038B2 · Trail · 2019 [cited by examiner]
US 10362296B2 · Price · 2019 [cited by examiner]
US 10469722B2 · Trail · 2019 [cited by examiner]
US 10694106B2 · John Archibald · 2020 [cited by examiner]
US 11010911B1 · Hall · 2021 [cited by examiner]
US 12033339B2 · Keinert · 2024 [cited by examiner]
US 20140368626A1 · John Archibald · 2014 [cited by examiner]
US 20140368688A1 · John Archibald · 2014 [cited by examiner]
US 20160253812A1 · Grossinger · 2016 [cited by examiner]
US 20220028103A1 · Hall et al. · 2022 [cited by applicant]
US 20240362805A1 · Nash · 2024 [cited by examiner]
International Search Report and Written Opinion—PCT/US2024/025497 —ISA/EPO—Jul. 19, 2024. [cited by applicant]