Squeezing bag-of-features for scalable and semantic 3D model retrieval
Squeezing bag-of-features for scalable and semantic 3D model retrieval
复制标题
DOI：
10.1109/cbmi.2010.5529890
复制
发表时间：
2010-06
期刊：
影响因子：
--
通讯作者：
Ryutarou Ohbuchi;M. Tezuka;T. Furuya;Takashi Oyobe
中科院分区：
文献类型：
--
作者：
Ryutarou Ohbuchi;M. Tezuka;T. Furuya;Takashi Oyobe
我们之前提出了一种用于基于形状的 3D 模型检索的多视图、密集采样、视觉特征袋算法 [2]。该方法对于中等大小的基准数据集（−1,000 个 3D 模型）（包括刚性和铰接的 3D 形状）实现了良好的检索性能。它也比具有类似检索性能的其他方法快得多。然而，该方法没有利用语义知识。我们希望检索结果能够反映多个（例如，-100）语义类别。此外，如果应用于更大的数据库（例如，-1M 模型），由于其特征向量大小（例如，30k 维度），通过数据库进行搜索可能会很昂贵。本文提出了一种通过将其长度特征向量投影到包含多个语义类的流形上来“压缩”其长度特征向量的方法。实验评估表明，在减少特征向量大小（例如从 30k 减少到小于 100）的同时，可以实现等于或优于原始特征的检索性能。
We have previously proposed a multiple-view, densely-sampled, bag-of-visual features algorithm for shape-based 3D model retrieval [2]. The method achieved good retrieval performance for moderately sized benchmark datasets (−1,000 3D models), including both rigid and articulated 3D shapes. It is also much faster than the other methods having similar retrieval performance. However, the method does not exploit semantic knowledge. We want the retrieval results to reflect multiple (e.g., −100) semantic classes. Also, if applied to a larger database (e.g., −1M models), search through the database can be expensive due to its large feature vector size (e.g., 30k dimensions). This paper proposes a method to "squeeze" its length feature vector by projecting it onto a manifold that incorporates multiple semantic classes. Experimental evaluation has shown that retrieval performances equal or better than original feature can be achieved while reducing feature vector size, e.g., from 30k down to less than 100.