A New Performance Benchmark for Content-based 3D Model Retrieval

摘要：

At first, the paper introduces the most prevailing 3D model benchmark, the Princeton Shape Benchmark.Deficiencies emerged in the benchmark are then discussed in depth, which are concluded as: 1) models belonging to the same category are not exactly similar according to their shapes, and 2) category similarity is totally ignored.To overcome those shortcomings, the paper proposes a new model classification method, based on which a novel retrieval performance metric, GSSS (Get Score from Similarity Sequence), is designed and discussed.Experimental results have shown that GSSS is better than the Precision-Recall benchmark on most occasions.

关键词： benchmark 3D Model Retrieval performance metric

作者: Jinjie Lin Yubin Yang Tong Lu Jiabin Ruan Wei Wei

作者单位: State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210093,Jiangsu,China State Key Laboratory for Novel Software Technology Nanjing University Nanjing 210093,Jiangsu,China;S

会议类型: 国际会议

会议名称: The Ninth International Conference on Web-Age Information Management(第九届web时代信息管理国际会议)(WAIM 2008)

会议地点: 张家界

会议语种:英文

在线出版日期: 2008-07-20（万方平台首次上网日期，不代表论文的发表时间）