In this paper we describe a novel 3D shape retrieval method based on Multi-scale Integral Orientations. In our approach, a 3D model after normalization is represented by a set of depth images captured uniformly on a unit circle. Then the shape descriptor based on the multiscale version of the localized gradient histogram is calculated for each depth image. Finally, the comparison is performed using a simple Euclidean distance to show the effectiveness of the shape descriptor for the 3D shape retrieval. We have then applied our algorithm on the Princeton shape benchmark and got results with performance similar to the Light Field Descriptor. In the future we are planning to use this descriptor with the bag of features and machine learning based approaches to get even better results.
Conference Dates: May 2, 2010
Conference Location: Norrköping, -1
Conference Title: Eurographics Workshop on 3D Object Retrieval (2010)
Pub Type: Conferences
3D shape retrieval, Multi-scale Integral Orientations, local features