Method and apparatus for distributed image data processing

Patent metadata

Field	Value
Publication number	US-2022406051-A1
Application number	US-202217845662-A
Country	US
Kind code	A1
Filing date	Jun 21, 2022
Priority date	Jun 22, 2021
Publication date	Dec 22, 2022
Grant date	—

Abstract

Disclosed herein are a method and apparatus for distributed image data processing. The method for distributed image data processing includes performing machine learning on an original image to produce a plurality of different task outputs, combining the plurality of task outputs to extract at least one final output, and compressing the final output and transmitting the final output to a server.

First claim

What is claimed is:

1 . A method for distributed image data processing, comprising:
performing machine learning on an original image to produce a plurality of different task outputs; combining the plurality of task outputs to extract at least one final output; and compressing the final output and transmitting the final output to a server.

2 . The method of claim 1 , wherein the task output comprises a first result, which is a description of a portion extracted from the original image, and a second result, which is an image containing a particular region extracted from the original image.

3 . The method of claim 2 , wherein the final output is extracted by respectively labeling the first results of the plurality of task outputs.

4 . The method of claim 2 , wherein the final output is extracted by combining overlapping regions of the second results of the plurality of task outputs.

5 . The method of claim 2 , wherein the first result comprises at least one of coordinates of the extracted portion, a color property of the extracted portion, or whether there is a pixel in the extracted portion or a combination thereof.

6 . The method of claim 1 , further comprising:
restoring the transmitted final output; and extracting an additional output by performing machine learning on the restored final output.

7 . The method of claim 6 , wherein the additional output is obtained by at least one of object tracking, pose estimation, action recognition, or a combination thereof.

8 . The method of claim 6 , further comprising:
providing the additional output to a user.

9 . The method of claim 1 , wherein the machine learning uses a neural network.

10 . An apparatus for distributed image data processing, comprising:
a memory configured to store a control program for distributed image data processing; and a processor configured to execute the control program stored in the memory, wherein the processor is configured to perform machine learning on an original image to produce a plurality of different task outputs, combine the plurality of task outputs to extract at least one final output, compress the final output, and transmit the final output to a server.

11 . The apparatus of claim 10 , wherein the task output comprises a first result, which is a description of a portion extracted from the original image, and a second result, which is an image containing a particular region extracted from the original image.

12 . The apparatus of claim 11 , wherein the final output is extracted by respectively labeling the first results of the plurality of task outputs.

13 . The apparatus of claim 11 , wherein the final output is extracted by combining overlapping regions of the second results of the plurality of task outputs.

14 . The apparatus of claim 11 , wherein the first result comprises at least one of coordinates of the extracted portion, a color property of the extracted portion, or whether there is a pixel in the extracted portion or a combination thereof.

15 . The apparatus of claim 10 , wherein the processor is further configured to restore the transmitted final output and extract an additional output by performing machine learning on the restored final output.

16 . The apparatus of claim 15 , wherein the additional output is obtained by at least one of object tracking, pose estimation, action recognition or a combination thereof.

17 . The apparatus of claim 15 , wherein the processor is further configured to provide the additional output to a user.

18 . The apparatus of claim 10 , wherein the machine learning uses a neural network.

Assignees

• Electronics & Telecommunications Res Inst
• Myongji Univ Industry And Academia Cooperation Foundation

Inventors

• Lee Jin-Young
• Lee Hee-Kyung
• Kim Sang-Kyun

Classifications

• G06V10/80

  Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level (multimodal speaker identification or verification G10L17/10) · CPC title

• G06V10/82 Primary

  using neural networks · CPC title

• G06V10/94

  Hardware or software architectures specially adapted for image or video understanding · CPC title

• G06V10/56

  relating to colour · CPC title

• G06V10/96 Primary

  Management of image or video recognition tasks · CPC title