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

Multi-quantization deep binary feature learning method and device

A technology of feature learning and quantization depth, which is applied in the field of computer vision and machine learning, can solve the problems of large-scale loss, achieve the effect of improving accuracy, meeting the needs of practical applications, and solving the problem of quantization error

Active Publication Date: 2020-01-17
TSINGHUA UNIV
View PDF3 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

However, the existing binary feature learning methods all use sign function for binarization, which leads to large quantization loss

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
  • Multi-quantization deep binary feature learning method and device
  • Multi-quantization deep binary feature learning method and device
  • Multi-quantization deep binary feature learning method and device

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0033] Embodiments of the present invention are described in detail below, examples of which are shown in the drawings, wherein the same or similar reference numerals designate the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by referring to the figures are exemplary and are intended to explain the present invention and should not be construed as limiting the present invention.

[0034] The following describes the multi-quantization deep binary feature learning method and device according to the embodiments of the present invention with reference to the accompanying drawings. First, the multi-quantization deep binary feature learning method according to the embodiments of the present invention will be described with reference to the accompanying drawings.

[0035] figure 1 It is a flow chart of the multi-quantization deep binary feature learning method of the embodiment of the present invention.

[0036] S...

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 method and device for learning multi-quantized deep binary features, wherein the method includes: extracting deep real-valued features of images; performing multi-quantization on the deep real-valued features of images through K self-encoding networks to obtain quantized Result; Binary encoding is performed on the deep real-valued features of the image according to the quantization results to obtain the binary features of the image. This method can effectively solve the quantization error problem caused by binarization, improve the accuracy of learning, and improve the learning efficiency, which is more efficient and simple, and better meets the needs of practical applications.

Description

technical field [0001] The invention relates to the technical field of computer vision and machine learning, in particular to a multi-quantized deep binary feature learning method and device. Background technique [0002] Visual recognition is a basic problem in the field of computer vision, which can be widely used in a variety of visual applications, such as face recognition, object recognition, scene recognition and texture recognition. As a classic pattern recognition problem, the main steps of visual recognition can be divided into: feature extraction and feature matching. The goal of feature representation is to obtain a feature vector for each picture, so that the feature vectors of similar pictures have a stronger similarity, and feature matching identifies the type of pictures based on the similarity measure of picture features. Due to the large differences in illumination, posture, background, viewing angle and occlusion of objects in the natural environment, the ...

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 Patents(China)
IPC IPC(8): G06K9/62G06K9/46G06K9/38
CPCG06V10/28G06V10/462G06F18/214
Inventor 鲁继文周杰段岳圻
Owner TSINGHUA UNIV