Granted Patent B2

US 12,548,162 · App. 18/189,459 · Granted Feb 10, 2026

Inventors: Niko Nevatie (Helsinki, FI); Erkki Parkkulainen (Helsinki, FI)

Assignee: TGI SPORT VIRTUAL TECHNOLOGIES LIMITED

G06T7/11G06T7/194G06T2207/20081G06T2207/20221G06T2207/30196G06T2207/30221

Abstract

Claims (37)

1 . A computer-implemented training method for training a segmentation model to segment an image; the method comprising receiving a training data set, the training data set comprising:

first image data representing a first image comprising a foreground and a background, the first image data being captured by at least one first visible electromagnetic radiation imaging device;

wherein the method comprises:

identifying at least one sub-section of the first image represented by the first image data, the identified sub-section corresponding to an image element identified in the foreground or the background of the first image;

augmenting the first image data by applying a first overlay image data to the at least one identified sub-section of the first image to generate first composite image data;

processing the first composite image data using the segmentation model;

updating the segmentation model in dependence on the processing of the first composite image data to generate a first updated segmentation model.

2 . A computer-implemented training method as claimed in claim 1 , wherein the method comprises:

augmenting the first image data by applying a second overlay image data to the first image to generate second composite image data;

processing the second composite image data using the first updated segmentation model; and

updating the first updated segmentation model in dependence on the processing of the second composite image to generate a second updated segmentation model.

3 . A computer-implemented training method as claimed in claim 2 , wherein augmenting the first image data comprises applying the second overlay image data to the at least one identified sub-section of the first image to generate the second composite image data.

4 . A computer-implemented training method as claimed in claim 1 , wherein the first training data sets each comprise first segmentation data for differentiating between the foreground and the background of the first image.

5 . A computer-implemented training method as claimed in claim 4 , wherein the first image comprises at least one background area occurring in the background of the first image, the first segmentation data identifying the at least one background area.

6 . A computer-implemented training method as claimed in claim 4 , wherein the first image comprises at least one foreground area occurring in the foreground of the first image, the first segmentation data identifying the at least one foreground area.

7 . A computer-implemented training method as claimed in claim 5 , wherein the or each sub-section of the first image identified in the first image corresponds to one of the at least one background area identified by the first segmentation data.

8 . A computer-implemented training method as claimed in claim 6 , wherein the or each sub-section of the first image identified in the first image corresponds to one of the at least one foreground area identified by the first segmentation data.

9 . A computer-implemented training method as claimed in claim 4 , wherein the first segmentation data is captured by at least one non-visible electromagnetic radiation imaging device.

10 . A non-transitory computer-readable medium having a set of instructions stored therein which, when executed, cause a processor to perform the method claimed in claim 1 .

11 . An image segmentation system for segmenting an image, the image segmentation system comprising one or more processors; wherein the one or more processors is configured to implement a segmentation model trained according to the method claimed in claim 1 .

12 . A content replacement system for replacing the content of an image, the content replacement system comprising an image segmentation system as claimed in claim 11 .

13 . An image processing system for processing an image, the image processing system comprising one or more processors having an electrical input for receiving first image data captured by at least one visible electromagnetic radiation imaging device and representing a first image comprising a foreground and a background;

wherein the one or more processors is configured to implement a segmentation model trained using the computer-implemented training method claimed in claim 1 ; the segmentation model being configured to segment the first image to differentiate between the foreground and the background of the first image.

14 . A computer-implemented method of processing a first image, the method comprising receiving first image data captured by at least one visible electromagnetic radiation imaging device and representing a first image comprising a foreground and a background;

wherein the method comprises implementing a segmentation model to segment the first image to differentiate between the foreground and the background of the first image, the segmentation model being trained using the computer-implemented training method claimed in claim 1 .

15 . A system for training a segmentation model to segment an image into one or more segments; the system comprising at least one processor and at least one memory device; the at least one processor comprising at least one input configured to receive a training data set comprising:

first image data representing a first image comprising a foreground and a background, the first image data being captured by at least one first visible electromagnetic radiation imaging device;

wherein, in respect of the or each training data set, the at least one processor is configured to:

identify at least one sub-section of the first image represented by the first image data, the identified sub-section corresponding to an image element identified in the foreground or the background of the first image;

augment the first image data by applying a first overlay image data to the at least one identified sub-section of the first image to generate first composite image data;

process the first composite image data using the segmentation model; and

update the segmentation model in dependence on the processing of the first composite image data to generate a first updated segmentation model.

16 . A system as claimed in claim 15 , wherein, in respect of the or each training data set, the at least one processor is configured to:

augment the first image data by applying a second overlay image data to the first image to generate second composite image data;

process the second composite image data using the first updated segmentation model; and

update the first updated segmentation model in dependence on the processing of the second composite image data.

17 . A system as claimed in claim 16 , wherein augmenting the first image data comprises applying the second overlay image data to the at least one identified sub-section of the first image to generate the second composite image data.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (81)