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

End-to-end image text translation method, system and device based on multi-task training

A multi-task and training sample technology, applied in natural language translation, neural learning methods, knowledge expression, etc., can solve the problems of lack of training data, model structure design, poor translation performance, etc., and achieve small storage space complexity and decoding time Less, improve the effect of training effect

Pending Publication Date: 2021-06-22
INST OF AUTOMATION CHINESE ACAD OF SCI
View PDF0 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] In order to solve the above-mentioned problems in the prior art, that is, in order to solve the problem of poor translation performance of the existing image-to-text translation model due to lack of training data and model structure design, the present invention proposes an end-to-end translation model based on multi-task training. A terminal image text translation method, the method comprising:

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
  • End-to-end image text translation method, system and device based on multi-task training
  • End-to-end image text translation method, system and device based on multi-task training
  • End-to-end image text translation method, system and device based on multi-task training

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0054] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments It is a part of embodiments of the present invention, but not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0055] 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 here are only used to explain related inventions, rather than to limit the invention. It should be noted that, in the case of no conflict, the embodime...

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 belongs to the technical field of natural language processing, particularly relates to an end-to-end image text translation method based on multi-task training, and aims to solve the problem that an existing image text translation model is poor in translation performance due to lack of training data and model structure design. The method comprises the following steps: acquiring data to be translated as input data; preprocessing the input data, and after preprocessing, inputting a pre-constructed image text translation model to obtain a translation result corresponding to the input data; wherein the image text translation model comprises a feature extractor and an encoder-decoder. According to the invention, image text translation performance is improved.

Description

technical field [0001] The invention belongs to the technical field of natural language processing, and in particular relates to an end-to-end image-to-text translation method, system, and device based on multi-task training. Background technique [0002] Image-to-text translation is the use of computer systems to automatically translate the source language contained in pictures or videos into the target language. Image-to-text translation technology can quickly and effectively help people translate and understand text content in pictures and videos. This technology can quickly translate the text of one language in images and videos to different languages ​​to facilitate the understanding of people who use different languages. [0003] The currently commonly used image-to-text translation architecture is to cascade the image-to-text recognition system with the machine translation system to translate the source language in the image. However, the two subtasks of the system ...

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
IPC IPC(8): G06F40/58G06F40/42G06N3/04G06N3/08G06N5/02
CPCG06F40/58G06F40/42G06N3/04G06N3/08G06N5/022Y02D10/00
Inventor 赵阳马聪张亚萍周玉
Owner INST OF AUTOMATION CHINESE ACAD OF SCI
Features
  • Generate Ideas
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com