A Visual Concept Detector and a Method for Constructing Semantic Fields

Abstract

The invention provides a visualization concept detector which is used for a retrieval process of a three-dimensional model. The visualization concept detector comprises a text description module, a semantic description module, a feature description module and a semantic concept computation module, wherein the text description module, the semantic description module and the feature description module are respectively used for constructing text description, network annotation and semantic concepts of synonym sets and shape and content characteristics in order to obtain a semantic concept vocabulary, and the text description, the network annotation and the semantic concepts are originated from a to-be-retrieved three-dimensional model. The semantic concept computation module is used for calculating through a probability model to obtain a core semantic concept vocabulary, namely a semantic field, and the visualization concept detector is combined to retrieve the three-dimensional model. Therefore, the harmful effect of 'semantic gap' is weakened, the calculated amount of three-dimensional model retrieval is reduced, the retrieval speed is accelerated, and the precision and the stability of retrieved results are enhanced.

Description

technical field [0001] The invention relates to the field of semantic-based three-dimensional model retrieval, in particular to a visual concept detector and a method for constructing a semantic field. Background technique [0002] As the amount of information increases geometrically, information retrieval is no longer limited to the single form of text retrieval, but has gradually expanded to multimedia retrieval and semantic retrieval. It has always been the goal of technicians to apply semantic-based retrieval methods in the multimedia field. [0003] At present, semantic-based 3D model retrieval methods are mainly divided into the following two types: 1. The user's relevant feedback information is regarded as a semantic concept. In this method, the user first submits a 3D model, and the system calculates the similarity between the 3D model to be retrieved and the shape features of the existing 3D models in the database, and returns the similar model to the user as the r...

AI Technical Summary

Problems solved by technology

[0004] In the above semantic-based 3D model retrieval method, there are the following problems: 1. The application of traditional semantic concepts in 3D model retrieval can easily cause a "semantic gap"

2. The traditional semantic concept construction method is only based on semantic annotation

The existing semantic concepts mainly come from semantic annotation, and there is a "semantic gap" problem between the high-level semantic concepts and the underlying 3D model; and the existing semantic concept construction methods also cause the user's semantic annotations to be too messy to correspond to accurate semantic concepts

These inaccurate semantic concepts will lead to low accuracy of the retrieval results of the 3D model

Images