IP Library › Granted Patent US 12664626

Granted Patent B2

US 12664626 · App. 18/473,166 · Granted Jun 23, 2026

Distortion correction for digital image sub-division

Inventors: Yanwei Wang (Bellevue, WA); Pascal Pare (Bothell, WA); Christopher Douglas Edmonds (Carnation, WA); Mark Anthony Plagge (Sammamish, WA); Miguel Susffalich (Kirkland, WA); Anuj Bharat Gosalia (Sammamish, WA)

Assignee: Microsoft Technology Licensing, LLC

G06T5/80 G06T5/10 G06T7/11 G06T17/00 G06V10/26 G06T2207/10032 G06T2207/20021 G06T2207/20048

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 12664626
App. No.: 18/473,166
Granted: Jun 23, 2026
Kind: B2

External Links

USPTO Patent Center↗ Full Patent PDF↗

Abstract

Digital image processing methods performed by a computer are disclosed. In one example, a digital image captured by a real camera having intrinsic and extrinsic parameters is received. One or more distortion correction transformations are applied to the digital image to generate a distortion-corrected digital image. The distortion-corrected digital image is sub-divided into a plurality of distortion-corrected sub-images. For each distortion-corrected sub-image of the plurality of distortion-corrected sub-images, the distortion-corrected sub-image is associated with a synthesized recapture camera having synthesized intrinsic and extrinsic parameters mapped from the intrinsic and extrinsic parameters of the real camera.

Claims (54)

1 . A digital image processing method performed by a computer, the method comprising:

receiving a digital image captured by a real camera having intrinsic and extrinsic parameters in which an intrinsic parameter includes a native principal point;

applying one or more distortion correction transformations to the digital image to generate a distortion-corrected digital image;

sub-dividing at least a portion of the distortion-corrected digital image into a plurality of distortion-corrected sub-images; and

for each distortion-corrected sub-image of the plurality of distortion-corrected sub-images:

mapping synthesized intrinsic and extrinsic parameters of a synthesized recapture camera for the distortion-corrected sub-image from the intrinsic and extrinsic parameters of the real camera in which a synthesized intrinsic parameter is mapped by redefining the native principal point with respect to a relative position of the distortion-corrected sub-image within the distortion-corrected digital image that differs from each other distortion-corrected sub-image of the plurality of distortion-corrected sub-images, and

within a storage subsystem of the computer, associating the distortion-corrected sub-image with the synthesized recapture camera having the synthesized intrinsic and extrinsic parameters mapped from the intrinsic and extrinsic parameters of the real camera.

2 . The method of claim 1 , wherein the native principal point is defined relative to an origin of a coordinate system of the distortion-corrected sub-image.

3 . The method of claim 2 , wherein the native principal point defined relative to the origin of the coordinate system of the distortion-corrected sub-image is outside the distortion-corrected sub-image.

4 . The method of claim 1 , wherein the digital image has a native spatial resolution, and wherein each distortion-corrected sub-image of the plurality of distortion-corrected sub-images maintains the native spatial resolution of the digital image.

5 . The method of claim 1 , wherein the one or more distortion correction transformations include at least one of a radial distortion correction transformation, a barrel distortion correction transformation and a pincushion distortion correction transformation.

6 . The method of claim 1 , further comprising:

identifying a target object in the distortion-corrected digital image; and

wherein the distortion-corrected digital image is sub-divided into the plurality of distortion-corrected sub-images based at least on the target object.

7 . The method of claim 6 , further comprising:

forming a bounding box around the target object in the distortion-corrected digital image; and

wherein each distortion-corrected sub-image of the plurality of distortion-corrected sub-images includes at least a portion of the bounding box.

8 . The method of claim 1 , further comprising:

identifying a target object in the distortion-corrected digital image; and

generating a three-dimensional (3D) model of the target object based at least on the plurality of distortion-corrected sub-images and the synthesized intrinsic and extrinsic parameters associated with the synthesized cameras corresponding to the plurality of distortion-corrected sub-images.

9 . The method of claim 8 , wherein the 3D model of the target object is generated using a Structure-from-Motion (SfM) algorithm.

10 . The method of claim 8 , wherein the 3D model of the target object is generated using a Multi-View Stereo (MVS) algorithm.

11 . A computing system comprising:

a logic subsystem; and

a storage subsystem holding instructions executable by the logic subsystem to:

receive a digital image captured by a real camera having intrinsic and extrinsic parameters in which an intrinsic parameter includes a native principal point;

apply one or more distortion correction transformations to the digital image to generate a distortion-corrected digital image;

sub-divide at least a portion of the distortion-corrected digital image into a plurality of distortion-corrected sub-images; and

for each distortion-corrected sub-image of the plurality of distortion-corrected sub-images:

map synthesized intrinsic and extrinsic parameters of a synthesized recapture camera for the distortion-corrected sub-image from the intrinsic and extrinsic parameters of the real camera in which a synthesized intrinsic parameter is mapped by redefining the native principal point with respect to a relative position of the distortion-corrected sub-image within the distortion-corrected digital image that differs from each other distortion-corrected sub-image of the plurality of distortion-corrected sub-images, and

within the storage subsystem of the computing system, associate the distortion-corrected sub-image with the synthesized recapture camera having the synthesized intrinsic and extrinsic parameters mapped from the intrinsic and extrinsic parameters of the real camera.

12 . The computing system of claim 11 , wherein the native principal point is defined relative to an origin of a coordinate system of the distortion-corrected sub-image.

13 . The computing system of claim 12 , wherein the native principal point defined relative to the origin of the coordinate system of the distortion-corrected sub-image is outside the distortion-corrected sub-image.

14 . The computing system of claim 11 , wherein the distortion-corrected digital image has a native spatial resolution, and wherein each distortion-corrected sub-image of the plurality of distortion-corrected sub-images maintains a same native spatial resolution as the distortion-corrected digital image.

15 . The computing system of claim 11 , wherein the one or more distortion correction transformations include at least one of a radial distortion correction transformation, a barrel distortion correction transformation and a pincushion distortion correction transformation.

16 . The computing system of claim 11 , wherein the storage subsystem holds instructions executable by the logic subsystem to:

identify a target object in the distortion-corrected digital image; and

wherein the distortion-corrected digital image is sub-divided into the plurality of distortion-corrected sub-images based at least on the target object.

17 . The computing system of claim 16 , wherein the storage subsystem holds instructions executable by the logic subsystem to:

form a bounding box around the target object in the distortion-corrected digital image; and

wherein each distortion-corrected sub-image of the plurality of distortion-corrected sub-images include at least a portion of the bounding box.

18 . The computing system of claim 11 , wherein the storage subsystem holds instructions executable by the logic subsystem to:

identify a target object in the distortion-corrected digital image; and

generate a three-dimensional (3D) model of the target object based at least on the plurality of distortion-corrected sub-images and the synthesized intrinsic and extrinsic parameters associated with the synthesized cameras corresponding to the plurality of distortion-corrected sub-images.

19 . A digital image processing method performed by a computer, the method comprising:

receiving a digital image captured by a real camera having intrinsic and extrinsic parameters in which an intrinsic parameter includes a native principal point;

applying one or more distortion correction transformations to the digital image to generate a distortion-corrected digital image;

identifying a target object in the distortion-corrected digital image;

sub-dividing at least a portion of the distortion-corrected digital image into a plurality of distortion-corrected sub-images; and

for each distortion-corrected sub-image of the plurality of distortion-corrected sub-images:

20 . The method of claim 19 , wherein the native principal point is defined relative to an origin of a coordinate system of the distortion-corrected sub-image.