Granted Patent B2

US 12,688,546 · App. 18/608,942 · Granted Jul 21, 2026

Inventor: Chia-Wei Hsiao (Hsinchu City, TW)

Assignee: Cupola360 INC.

G06V10/25G06T7/11G06T11/60

Abstract

Claims (38)

1 . An image processing device for rendering images, comprising:

a transceiver, configured to receive layout information of an output image and a vertex list corresponding to an input image;

a storage medium, configured to store a plurality of modules; and

a processing device, comprising a first processing unit and a second processing unit, wherein the processing device is coupled to the storage medium and the transceiver, and accesses and executes the plurality of modules, and the plurality of modules comprise:

a real-time composer, configured to:

divide a layout in the layout information according to a first computational capacity of the first processing unit and a second computational capacity of the second processing unit to obtain a first region of interest corresponding to the first processing unit and a second region of interest corresponding to the second processing unit;

obtain a first data structure corresponding to a first vertex of the first region of interest and a second data structure corresponding to a second vertex of the second region of interest from the vertex list;

perform a first interpolation operation according to the first data structure using the first processing unit to generate a third data structure corresponding to a third vertex of the first region of interest;

perform a second interpolation operation according to the second data structure using the second processing unit to generate a fourth data structure corresponding to a fourth vertex of the second region of interest; and

update the vertex list according to the third data structure and the fourth data structure to generate a composed vertex list; and

an image processing module, configured to map the input image to the layout according to the composed vertex list to generate the output image, and output the output image through the transceiver.

2 . The image processing device according to claim 1 , wherein the real-time composer is further configured to:

calculate a first weight of the first processing unit and a second weight of the second processing unit according to the first computational capacity and the second computational capacity, wherein the first weight is proportional to the first computational capacity;

calculate a first height corresponding to the first region of interest according to the first weight and the second weight; and

divide the layout according to the first height to obtain the first region of interest.

3 . The image processing device according to claim 2 , wherein the real-time composer is further configured to:

updated the first weight according to an execution time of the first interpolation operation.

4 . The image processing device according to claim 3 , wherein the first interpolation operation corresponds to a current frame of the input image, and the real-time composer is further configured to:

input a ratio of the first height to the execution time and a third weight corresponding to a previous frame of the input image to an infinite impulse response filter to update the first weight, wherein the third weight corresponds to the first processing unit.

5 . The image processing device according to claim 1 , wherein the real-time composer is further configured to:

store a plurality of data structures corresponding to a plurality of vertices of the first region of interest in a first storage area of the storage medium, wherein the plurality of data structures comprise the first data structure;

generate a first address table corresponding to the first storage area; and

perform the first interpolation operation according to the first address table using the first processing unit to generate the third data structure, and store the third data structure in the first storage area.

6 . The image processing device according to claim 5 , wherein the real-time composer is further configured to:

combine the first address table corresponding to the first storage area and a second address table corresponding to a second storage area in the storage medium to generate the composed vertex list, wherein the second storage area corresponds to the second region of interest.

7 . The image processing device according to claim 1 , wherein the layout information comprises coordinates, a width, and a height corresponding to a region of interest of a source image and comprises coordinates, a width, and a height corresponding to a display area of a target image.

8 . The image processing device according to claim 1 , wherein the real-time composer is further configured to:

obtain a correction map corresponding to the input image through the transceiver, wherein the vertex list comprises the first vertex on the correction map;

divide the first region of interest into a plurality of tiles comprising a first tile, and map a sub-block of the first tile to the correction map to obtain a mapping area; and

in response to the third vertex in the mapping area corresponding to the first vertex, perform the first interpolation operation according to the first data structure to generate the third data structure.

9 . An image processing method for rendering images, comprising:

receiving layout information of an output image and a vertex list corresponding to an input image;

dividing a layout in the layout information according to a first computational capacity of a first processing unit and a second computational capacity of a second processing unit to obtain a first region of interest corresponding to the first processing unit and a second region of interest corresponding to the second processing unit;

obtaining a first data structure corresponding to a first vertex of the first region of interest and a second data structure corresponding to a second vertex of the second region of interest from the vertex list;

performing a first interpolation operation according to the first data structure by the first processing unit to generate a third data structure corresponding to a third vertex of the first region of interest;

performing a second interpolation operation according to the second data structure by the second processing unit to generate a fourth data structure corresponding to a fourth vertex of the second region of interest;

updating the vertex list according to the third data structure and the fourth data structure to generate a composed vertex list; and

mapping the input image to the layout according to the composed vertex list to generate the output image, and outputting the output image.

Assignments (2)

Priority Claims (1)

Continuity (1)

References Cited (15)