Article

Article title 3D-MODEL NORMALIZATION FOR INTEGRAL SPIN IMAGE ESTIMATION
Authors А.А. Krylovetsky, I.S. Chernikov
Section SECTION IV. MATHEMATICAL METHODS OF AN ARTIFICIAL INTELLECT
Month, Year 06, 2012 @en
Index UDC 004.9
DOI
Abstract 3d-model global surface descriptor construction is one of the most complex issues in three-dimensional computer vision. Global surface descriptors describe the shape of the whole model’s surface and are basically used for model matching. New global descriptors are based on the concept of popular local surface descriptors spin images. Due to its easy computation and effective descriptive features the global surface descriptors show sufficiently good results in model matching. The requirement of special preliminary 3d-model normalization is the peculiarity of new global descriptors. In this paper we present effective and robust model normalization method and introduce the idea of new global surface descriptors integral spin images.

Download PDF

Keywords 3d-model; normalization; global surface descriptor; integral spin image.
References 1. Kazhdan M. Shape Representations And Algorithms For 3D-Model Retrieval. PhD thesis,Princeton University, 2004.
2. Johnson A.E., Hebert M. IEEE Trans // Pattern Analysis and Machine Intelligence. – 1999.– № 21 (5). – P. 433-449.
3. Крыловецкий А.А., Черников И.С. Модифицированные спиновые изображения в системах трехмерной реконструкции // Телематика'2010: Тр. XVII Всерос. науч.-метод. конф.– СПб. 21-24 июня 2010 г.
4. Princeton Shape Benchmark, http://shape.cs.princeton.edu/benchmark/.

Comments are closed.