Login| Sign Up| Help| Contact|

Patent Searching and Data


Title:
CLOTHES LOGO EFFECT DETECTION METHOD AND APPARATUS, AND ELECTRONIC DEVICE
Document Type and Number:
WIPO Patent Application WO/2019/096179
Kind Code:
A1
Abstract:
The present invention relates to the technical field of image identifications, and provides a clothes LOGO effect detection method and apparatus, and an electronic device. The clothes LOGO effect detection method comprises: obtaining a clothes LOGO image to be detected (S101); extracting LOGO characteristics of the clothes LOGO image to be detected according to a preset deep learning model, the preset deep learning model comprising a deep learning model based on an adversarial neural network (S102); and using a classifier based on preset deep learning model training to identify the LOGO characteristics, and generating the identification result of the clothes LOGO image to be detected, the identification result comprising effect pictures corresponding to a plurality of techniques (S103). The method can identify the clothes LOGO images by means of the deep learning model based on the adversarial neural network to generate effect pictures under different techniques, so as to enable the work staff to take reference without drawing a design for the clothes LOGO, thus saving plenty of time and cost.

Inventors:
SCOTT MATTHEW ROBERT (CN)
HUANG DINGLONG (CN)
FU KAI (CN)
Application Number:
PCT/CN2018/115496
Publication Date:
May 23, 2019
Filing Date:
November 14, 2018
Export Citation:
Click for automatic bibliography generation   Help
Assignee:
SHENZHEN MALONG TECH CO LTD (CN)
International Classes:
G06T7/00
Foreign References:
CN107845092A2018-03-27
CN105631461A2016-06-01
US20170316281A12017-11-02
US20160342859A12016-11-24
CN101782771A2010-07-21
CN107204025A2017-09-26
Attorney, Agent or Firm:
CHOFN INTELLECTUAL PROPERTY (CN)
Download PDF: