Cross-modal retrieval method and device based on hash algorithm and neighborhood graph
A hash algorithm and cross-modal technology, applied in the field of retrieval, can solve the problems of not considering the similarity between modalities, poor cross-modal retrieval effect, etc., and achieve the effect of improving comprehensiveness and accuracy
Pending Publication Date: 2021-01-08
GCI SCI & TECH +1
View PDF6 Cites 2 Cited by
- Summary
- Abstract
- Description
- Claims
- Application Information
AI Technical Summary
Problems solved by technology
[0003] The present invention provides a cross-modal retrieval method and device based on a hash algorithm and a neighborhood graph to solve the problem that the existing cross-modal retrieval method does not consider the similarity between samples in the same modality and the similarity, resulting in poor performance of cross-modal retrieval technical problems
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
[0031] The following will clearly and completely describe the technical solutions in the embodiments of the application with reference to the drawings in the embodiments of the application. Apparently, the described embodiments are only some of the embodiments of the application, not all of them. Based on the embodiments in this application, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the scope of protection of this application.
[0032] In the description of the present application, it should be understood that the terms "first" and "second" are used for description purposes only, and cannot be interpreted as indicating or implying relative importance or implicitly indicating the quantity of indicated technical features. Thus, a feature defined as "first" and "second" may explicitly or implicitly include one or more of these features. In the description of the present application, unless otherwise specified,...
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 invention discloses a cross-modal retrieval method and device based on a hash algorithm and a neighborhood graph, and the method comprises the steps: obtaining a multi-modal original sample, minimizing a residual value obtained before and after feature transformation of the multi-modal original sample to acquire a minimized residual value; learning potential association among the multi-modal original samples according to a collaborative matrix decomposition method, and calculating according to the potential association to obtain semantic consistency among modals of the multi-modal originalsamples; adopting popular learning of a neighborhood graph to calculate and obtain semantic consistency in modals of the multi-modal original sample; and minimizing the residual value, the semantic consistency between the modals and the semantic consistency in the modals, and combining regularization calculation avoiding overfitting to obtain an objective function. According to the embodiment ofthe invention, the target function for cross-modal retrieval is calculated by comprehensively considering the global features of multiple modals and the local features between the modals, so that thecomprehensiveness and accuracy of cross-modal retrieval are improved.
Description
technical field [0001] The invention relates to the technical field of retrieval, in particular to a cross-modal retrieval method and device based on a hash algorithm and a neighborhood graph. Background technique [0002] The rapid development of information technology has brought about the explosive growth of multimodal data, including multi-source heterogeneous data such as images, audio, text, and video. Due to the heterogeneous differences in semantic representation between modalities, efficient multimodal retrieval has become one of the key issues in current multimodal fusion. In the prior art, most of the multimodal retrieval is realized by hash algorithm. The hash algorithm maps multimodal data to a unified latent space, and the hash code obtained by quantizing the feature vector through the hash function realizes multimodal retrieval. The alignment of the modal space. However, the applicant found in the research that the existing cross-modal retrieval methods do n...
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 Applications(China)
IPC IPC(8): G06F16/53G06K9/62
CPCG06F16/53G06F18/22G06F18/214
Inventor 杜翠凤蒋仕宝孙广波朱春荣
Owner GCI SCI & TECH