Unlock instant, AI-driven research and patent intelligence for your innovation.

Neural network operation method using grid generator and apparatus using same

A neural network and grid generation technology, applied in biological neural network models, neural learning methods, probability networks, etc., can solve the problems of low CNN operation efficiency and inability to calculate test images correctly.

Active Publication Date: 2020-07-28
STRADVISION
View PDF6 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] However, the learning process described above has a significant disadvantage
That is, when the arrangement of the input image is different from the typical arrangement of the training images, the CNN operation is inefficient
For example, in the case of a turn, unlike training images with a typical arrangement, there is no lane in the middle of the test image, since the parameters of the CNN are optimized for input images where the lane is in the middle of the image, the test image cannot be correctly calculated using the above parameters

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Neural network operation method using grid generator and apparatus using same
  • Neural network operation method using grid generator and apparatus using same
  • Neural network operation method using grid generator and apparatus using same

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0055] figure 1 It is a diagram showing the configuration of a computing device for executing a neural network calculation method using a mesh generator according to the present invention.

[0056] refer to figure 1 , the computing device 100 may include a detector 130 , a grid generator 140 and a neural network 150 . Various data input / output processing and calculation processing of the detector 130, the mesh generator 140, and the neural network 150 can be performed by the communication section 110 and the processor 120, respectively. However, in figure 1 In , a detailed description of the connection relationship between the communication unit 110 and the processor 120 is omitted. In addition, the computing device 100 may further include a memory 115 capable of storing computer-readable instructions for performing processes described later. As one example, processors, memory, media, etc. may be integrated into one processor to function.

[0057] As described above, the ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a neural network operation method using a grid generator and an apparatus using the same. The method includes steps: a computing device (a) instructing a detector to acquire object location information for testing and class information; (b) instructing the grid generator to generate section information by referring to the object location information for testing; (c) instructing a neural network to determine parameters for testing, to be used for applying the neural network operations to either (i) the subsections including each of the objects for testing and each of non-objects for testing, or (ii) each of sub-regions, in each of the subsections, where said each of the non-objects for testing is located; and (d) instructing the neural network to apply the neural network operations to the test image for testing to thereby generate neural network outputs.

Description

technical field [0001] The present invention relates to a neural network operation method using a grid generator and a device using the method in order to switch modes according to the category of an area in a test image to satisfy Level 4 of an autonomous vehicle, and specifically, to use the grid generator A neural network operation method for a device, comprising the following steps: (a) if a test image is obtained, the detector is made to detect the test object and the non-object for test existing in the test image, and obtain the position information and the location information of the test object The category information of the non-object for testing, the position information of the object for testing includes information related to the position of the object for testing on the test image, the category information of the non-object for testing includes information related to the presence information related to the category of the test non-object in the test image; (b) ca...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06K9/00G06K9/34G06K9/62G06N3/02G06V10/764
CPCG06N3/02G06V20/56G06V10/267G06F18/214G06N3/08G06V10/82G06V10/764G06N3/045G06N3/04G06N3/063G06T7/70G06T2207/30252G06F18/217G06F18/24G06F18/2148G06F18/2163G06F18/2431G06N7/01
Inventor 金桂贤金镕重金寅洙金鹤京南云铉夫硕焄成明哲吕东勋柳宇宙张泰雄郑景中诸泓模赵浩辰
Owner STRADVISION