Furniture model retrieval method based on voice and sketch

Abstract

The invention belongs to the technical field of furniture model retrieval in computer aided design, and particularly relates to a furniture model retrieval method based on voice and sketch, which comprises the following steps: S1, constructing a database based on a neural network, standard keyword description and model description semantic features; s2, inputting the two-dimensional sketch into a database to obtain a sketch retrieval result; s3, inputting voice into a database to obtain a voice retrieval result; and S4, based on the sketch retrieval result and the voice retrieval result, obtaining a final retrieval result in a voting mode. According to the rapid and accurate furniture model retrieval method designed by the invention, a large amount of design cost and time expenditure are saved, so that the utilization rate and the use value of the furniture model are improved.

Description

technical field [0001] The invention belongs to the technical field of retrieval of furniture models in computer-aided design, and in particular relates to a retrieval method of furniture models based on speech and sketches. Background technique [0002] With the development of computer-aided design, furniture design is more and more dependent on computer-aided design. In the design of home scenes, the retrieval of furniture models is a key issue. Due to the large number of furniture types and models, home designers need a fast and accurate way to retrieve furniture models. The furniture retrieval task mainly includes two aspects: extracting model features and establishing the relationship between input and model features. [0003] The retrieval usually uses the following retrieval methods: text-based retrieval and sketch-based retrieval. The method based on text retrieval requires a large number of manual annotations on the model, and the model features are stored in the...

AI Technical Summary

Problems solved by technology

[0005] For example, the Chinese patent application number is: CN201810597066.1 discloses a sketch-based 3D model retrieval method, which processes the 3D model into multiple perspective screenshots, and then uses different contour extraction operators to process it into different styles of sketches , get the sketch data set, label the data set and use it in the training and testing process; on the basis of the convolutional neural network, a layered network method is added, a large classification network is used to train a large classification, and the input sketch Classified into 40 major categories, 40 small classification networks train the specific differences in each category of models, and classify the input sketches to a specific 3D model in a certain category, which can be retrieved with high accuracy. The three-dimensional model has little information redundancy. The invention uses the angle matrix calculated by sampling contour points as the input of the convolutional neural network, which is more in line with the characteristics of sketches and has high retrieval accuracy. However, a furniture based on voice and sketches is not proposed. Model Retrieval Method

Images