Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

A clothing attribute tag identification method based on fine classification of a neural network model

A neural network model and fine classification technology, applied in biological neural network models, neural learning methods, character and pattern recognition, etc., to achieve the effect of improving the accurate recognition rate and reducing the influence of background factors

Inactive Publication Date: 2019-05-03
TIANJIN UNIV
View PDF2 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

For image classification, there are also great technical challenges. Specifically, there are viewpoint changes, scale changes, intra-class changes, image deformation, image occlusion, lighting conditions and background noise, etc.

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
  • A clothing attribute tag identification method based on fine classification of a neural network model
  • A clothing attribute tag identification method based on fine classification of a neural network model
  • A clothing attribute tag identification method based on fine classification of a neural network model

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0024] The present invention will be further described below in conjunction with the accompanying drawings.

[0025] Starting from multiple dimensions, the present invention can list all the attributes of the clothing. From the characteristics of the clothing, three perspectives can be obtained, namely the upper body perspective, the lower body perspective, and the whole body perspective. As far as the upper body perspective is concerned, attributes can be obtained including: collar, sleeves, and clothing. For the lower body perspective, the attributes that can be obtained include skirts and pants. For a full-body view, silhouette and waist attributes can be obtained. After fine-scoring the model, more refined attributes can be obtained. Taking the collar as an example, you will get neck design, collar design, lapel design, neckline design, and detail design. Each of these designs contains many different design types. In this embodiment, the neck design is taken as an exam...

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 clothing attribute tag identification method based on fine classification of a neural network model, which comprises the following steps: (1) cutting a collected clothing picture by using a target detection algorithm Faster RCNN, and removing redundant irrelevant backgrounds; (2) separating different background type pictures into a tiled picture and a model picture; (3) carrying out data augmentation on the obtained picture, including reflection, rotation and random stretching operation, increasing the number of pictures, and obtaining augmented training data; (4) taking the obtained augmented training data as a training set, training a NasNet improved multi-classification network, and training an inception resnetv2 improved multi-classification network; and (5) fusing the network models trained by NasNet and inception netv2, and outputting the attribute tags of the clothes by adopting a weighted average fusion method. The method can be widely applied to application scenarios such as clothing image retrieval, tag navigation and clothing matching.

Description

technical field [0001] The invention relates to computer vision processing based on a neural network model to realize image classification and target detection, and in particular to a clothing attribute label recognition method based on fine classification of a neural network model. Background technique [0002] With the rapid development of computer vision (machine vision), object classification and target detection methods in computer vision have been applied to all aspects of real life, such as automatic driving, facial recognition payment, industrial automatic classification machine, image retrieval, etc. Applications. Recent advances in neural networks and deep learning have greatly advanced the development of these state-of-the-art visual recognition systems. And computer vision has also shown its prominence in the fashion industry. AI fashion has made greater changes to our lives and has had a huge impact on our dressing and matching. The classification of clothing a...

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): G06K9/00G06K9/62G06N3/04G06N3/08
Inventor 吴昊葛卫民
Owner TIANJIN UNIV
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products