Construction method of reward and image description model in image description model
An image description and construction method technology, applied in the field of reward and image description model construction, can solve the problem of low image description accuracy, and achieve the effect of solving the problem of uneven word distribution, improving accuracy, and reducing relevance
Active Publication Date: 2021-03-30
暗物智能科技(广州)有限公司
View PDF10 Cites 0 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0006] In view of this, the embodiment of the present invention provides a reward and image description model construction method in the image description model to solve the problem of low accuracy of image description
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 moreImage
Smart Image Click on the blue labels to locate them in the text.
Smart ImageViewing Examples
Examples
Experimental program
Comparison scheme
Effect test
Embodiment Construction
[0078] 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 those skilled in the art without making creative efforts belong to the protection scope of the present invention.
[0079] In order to clearly describe the embodiments of the present invention, some nouns involved in the embodiments of the present invention are explained as follows:
[0080] (1) Training set: a collection of several images, each image corresponds to a certain number of label descriptions; for example, each image in the training set corresponds to 5 la...
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
Login to View More Abstract
The present invention relates to the technical field of image description, and in particular to a method for constructing a reward and an image description model in an image description model, wherein the method for constructing a reward includes: obtaining a label description of a test image and generating a description based on a preset image description model based on the test image; Tag description and generated description, calculate the reward of the n-tuple in the generated description; compare the size relationship between the reward of the n-tuple and the first threshold and the size relationship between the reward of each word in the n-tuple and the second threshold to filter out Generate the word to be promoted in the description; calculate the reward for the word to be promoted according to the label description and the reward of the n-tuple corresponding to the word to be promoted. Implementing its constraints by giving higher rewards to words that appear less frequently can well solve the problem of uneven word distribution and help capture more specific visual details of images to improve the accuracy of image descriptions .
Description
technical field [0001] The invention relates to the technical field of image description, in particular to a reward in an image description model and a construction method of an image description model. Background technique [0002] The task of image description is a common task in everyday life. Its purpose is to generate a grammatical natural language description describing the content of the image based on the content of the image. This task can be applied in early childhood education, navigation and guidance in real life. [0003] The intersection of computer vision and natural language processing is the two major fields of artificial intelligence. Although both fields use similar artificial intelligence and machine learning methods for field research, there is still little interaction between the two fields. In recent years, however, the fields of natural language processing and computer vision have made enormous strides in analyzing and generating text and understand...
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
Login to View More Patent Type & Authority Patents(China)
IPC IPC(8): G06K9/62G06N3/04G06N3/08G06F16/583
CPCG06N3/08G06F16/583G06N3/045G06F18/22
Inventor 陈添水吴捷梁小丹林倞
Owner 暗物智能科技(广州)有限公司