摘要
提出了一种基于代表性视图的三维模型检索方法。在三维模型的视图表示方面,为了充分表示模型,并减少冗余信息,首先采用光场描述符(light field descriptor,LFD)将三维模型投影成二维视图,再将二维视图采用k均值聚类算法(K-means clustering algorithm,K-MEANS)进行聚类,生成代表性视图。然后采用卷积神经网络(convolutional neural network,CNN)提取视图特征并进行分类。同时提出了一种支持多种查询方式的相似度评价方法,以实现草图、图片或三维模型为输入条件的模型检索。本文在ModelNet40模型库上的实验结果表明,部分特征突出的三维模型检索的准确率可以达到100%。
3D model retrieval based on representative views was proposed.On the view representation of the 3D model,in order to fully represent the model and reduce redundant information,we firstly adopt Light Field Descriptor(LFD)to generate 2D views,and then use K-MEANS to get representative views from the 2D views.Next,a Convolution Neural Network(CNN)is adopted to extract the view feature and classify.At the same time,a similarity metrics supporting multiple query method is proposed to realize model retrieval with sketches,pictures or 3D models as input.Results on ModelNet40 showed that the proposed method could achieve an accuracy of 100%for part of models with distinct features.
作者
丁博
汤磊
何勇军
于军
DING Bo;TANG Lei;HE Yong-jun;YU Jun(School of Computer Science and Technology, Harbin University of Science and Technology, Harbin 150080, China;School of Automation, Harbin University of Science and Technology, Harbin 150080, China)
出处
《哈尔滨理工大学学报》
CAS
北大核心
2021年第6期18-23,共6页
Journal of Harbin University of Science and Technology
基金
国家自然科学基金面上项目(61673142)
黑龙江省自然科学基金杰出青年项目(JJ2019JQ0013)
黑龙江省普通本科高等学校青年创新人才项目(UNPYSCT-2016034).
