Method and device for searching coordinate recognition model based on NAS technology

A coordinate recognition and baseline model technology, applied in the field of computer vision, can solve the problems of reducing the accuracy and speed of neural network models, lack of targeted optimization of human action categories, etc., and achieve the effects of reducing performance requirements, saving time, and improving accuracy and speed
CN112861689APending Publication Date: 2021-05-28SHANGHAI YITU NETWORK SCI & TECH
0 Cites 1 Cited by

Patent Information

Authority / Receiving Office
CN · China
Current Assignee / Owner
SHANGHAI YITU NETWORK SCI & TECH
Publication Date
2021-05-28

Smart Images

  • Figure 1
    Figure 1
  • Figure 2
    Figure 2
  • Figure 3
    Figure 3
Patent Text Reader

Abstract

The invention relates to the technical field of computer vision, and relates to a method and device for searching a coordinate recognition model based on an NAS technology, in particular to a method and device for training the coordinate recognition model, and the method comprises the steps: obtaining the value range of each hyper-parameter of a baseline model, and obtaining the value range of each hyper-parameter according to the value range of each hyper-parameter; combining the hyper-parameters under different values to generate a plurality of numerical value combinations; for each numerical value combination, setting each hyper-parameter of the baseline model as each value in any numerical value combination, and obtaining a candidate coordinate identification model under the numerical value combination; for each candidate coordinate recognition model, inputting the image sample set into any candidate coordinate recognition model for training, and calculating an error value of the candidate coordinate recognition model; and taking the candidate coordinate recognition model meeting the preset error value condition as a final optimized coordinate recognition model, so that automatic design of the coordinate recognition model can be realized in combination with NAS.
Need to check novelty before this filing date? Find Prior Art

Description

technical field

[0001] The present application relates to the technical field of computer vision, and in particular to a search method and device for a coordinate recognition model based on NAS technology. The present application specifically provides a method and device for training a coordinate recognition model. Background technique

[0002] At present, when the action category of the human body contained in the image to be recognized is recognized, it is generally realized through a neural network model manually designed and trained in advance. However, due to the lack of targeted optimization of the human action category through the manual design of the model, the accuracy and speed of the neural network model in recognizing human actions will be reduced.

[0003] In order to solve the above problems, in the related art, the network structure contained in the neural network model can be automatically designed through neural network search (Neural Architecture Search, N...

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