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

Object classifying and model training method, device, medium and system

A training method and a technology for classifying models, which are applied in character and pattern recognition, instruments, computer components, etc., can solve the problems of increased difficulty in model training, and achieve the effect of reducing resource burden and simplifying complexity

Pending Publication Date: 2018-06-29
BEIJING SENSETIME TECH DEV CO LTD
View PDF5 Cites 2 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For example, the difficulty of training a model used to perform classification increases

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
  • Object classifying and model training method, device, medium and system
  • Object classifying and model training method, device, medium and system
  • Object classifying and model training method, device, medium and system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0036] The application will be further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application, but not to limit the present application. In addition, it should be noted that, for ease of description, only parts relevant to the present application are shown in the drawings. The present application will be described in detail below with reference to the accompanying drawings and embodiments. It should be understood that unless otherwise stated, ordinal numbers, such as "first", "second", etc., used herein are only used to distinguish one element from another and do not indicate importance or priority. For example, the first weight and the second weight represent different weights.

[0037] It should be noted that, in the case of no conflict, the embodiments in the present application and the features in the embodiments can be...

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 provides object classifying and model training method, device, medium and system. An object classifying model comprises a feature extractor and a classifier. The training method comprises the following steps: extracting, by the feature extractor, the features of an object from data items including the object; performing, by the classifier, down-sampling of partial classes from a predefined candidate class set, acquiring response of the extracted features to the down-sampled partial classes, and classifying the object based on the response; and adjusting parameters in the classifier and / or feature extractor according to the classification result.

Description

technical field [0001] The present application relates to the field of artificial intelligence (AI), and more specifically, to an object classification and model training method, device, medium and system. Background technique [0002] In recent years, thanks to the development of deep learning, many technological breakthroughs have occurred in the field of AI. The field of AI has many branches and development directions, but the problem of object classification is always a core problem. [0003] In object classification problems, some candidate categories are often predefined for objects to be classified. When the number of candidate categories increases, the difficulty of classification increases. For example, the training difficulty of models used to perform classification increases. When the number of candidate categories increases to a certain level (for example, hundreds of thousands or even millions of levels), the object classification problem evolves into a massi...

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): G06K9/62
CPCG06F18/2413G06F18/214
Inventor 张行程杨磊林达华
Owner BEIJING SENSETIME TECH DEV CO LTD