Eureka AIR delivers breakthrough ideas for toughest innovation challenges, trusted by R&D personnel around the world.

Retrieval model training method and device, retrieval method and device, electronic equipment and storage medium

A model training and three-dimensional model technology, applied in neural learning methods, biological neural network models, digital data information retrieval, etc., can solve the problems of category information interference color and shape features, difficult sample mining, coupling, etc., to reduce interference , Improve the convergence speed and improve the accuracy

Pending Publication Date: 2022-01-21
INST OF COMPUTING TECH CHINESE ACAD OF SCI
View PDF0 Cites 0 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0004] Aiming at the deficiencies of the prior art, the main purpose of the present invention is to propose a 3D model retrieval model training method and device, a 3D model retrieval method and device, electronic equipment and storage media, which can overcome the difficulties in image-based 3D model retrieval tasks. Problems such as sample mining, interference of category information on retrieval, and coupling of color and shape features

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
  • Retrieval model training method and device, retrieval method and device, electronic equipment and storage medium
  • Retrieval model training method and device, retrieval method and device, electronic equipment and storage medium
  • Retrieval model training method and device, retrieval method and device, electronic equipment and storage medium

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0066] In order to make the above-mentioned features and effects of the present invention more clear and understandable, the following specific examples are given together with the accompanying drawings for detailed description as follows.

[0067] In view of the various limitations of existing methods, the present invention intends to improve the entire retrieval algorithm from the perspectives of supervision items and data enhancement. In the present invention, one focus is to construct a double-contrastive loss function between instances and categories to replace the triplet loss. As a popular research direction in recent years, contrastive learning uses contrastive loss, so that different data enhancements of the same sample have similar codes, and different samples have different codes, which has made great progress in unsupervised representation learning. success. Specifically in this work, the mechanism of contrastive learning coincides with the method of metric learni...

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 provides a three-dimensional model retrieval model training method. The method comprises the following steps: rendering a three-dimensional model in a three-dimensional model database to obtain a corresponding multi-view rendering grayscale image; in the trained query image small batch, performing color conversion on the query image to obtain a corresponding data enhancement image; sending the data enhancement image and the corresponding mask image into a query image encoder to obtain a query code corresponding to the query image; sending the multi-view rendering grayscale image corresponding to the three-dimensional model into a rendering encoder to obtain a rendering code corresponding to the multi-view rendering grayscale image corresponding to the three-dimensional model; sending the query code and the rendering code into an attention mechanism module to obtain a specific code of the three-dimensional model for the current query image; and optimizing by utilizing the loss function to obtain a three-dimensional model retrieval model.

Description

technical field [0001] The invention relates to the geometric processing field of computer graphics, in particular to a three-dimensional model retrieval model training method and device, a three-dimensional model retrieval method and device, electronic equipment and a storage medium. Background technique [0002] Multimedia retrieval, including image retrieval and 3D model retrieval, has always been a very basic and important frontier hot issue in the field of computer vision and graphics. One of its research challenges is how to characterize the similarity relationship between retrieved images of different modalities and 3D models. Due to the development of deep learning and 3D model datasets with rich objects and categories, as well as its wide range of applications, including scene reconstruction, 3D printing, virtual reality and e-commerce platforms, the task of 3D model retrieval based on a single real image has recently gained a lot of attention. more attention. In ...

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): G06F16/583G06F16/532G06N3/08
CPCG06F16/583G06F16/532G06N3/08
Inventor 高林林明仙杨洁
Owner INST OF COMPUTING TECH CHINESE ACAD OF SCI
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Eureka Blog
Learn More
PatSnap group products