IIIT Hyderabad Publications |
|||||||||
|
SplineNet: B-spline neural network for efficient classification of 3D dataAuthors: Sai Sagar Jinka,Avinash Sharma Conference: 11th Indian Conference on Computer Vision, Graphics and Image Processing (ICVGIP-2018 2018) Location Hyderabad, India Date: 2018-12-18 Report no: IIIT/TR/2018/59 AbstractAdvancement in the field of 3D capture, owing to use of consumer depth sensors, has reinvigorated the research interest for scalable shape classification and recognition algorithms. Majority of recent deep learning pipelines for 3D shapes uses volumetric representation, extending the concept of 2D convolution to 3D domain. Nevertheless, the volumetric representation poses a serious computational disadvantage as most of the voxel grids are empty and results in redundant computation. Moreover, a 3D shape is determined by its surface and hence performing convolutions on the voxels inside the shape is sheer wastage of computation. In this paper, we focus on constructing a novel, fast and robust characterization of 3D shapes that accounts for local geometric variations as well as global structure. We built up on the learning scheme of [19] by introducing sets of B-spline surfaces instead of point filters, in order to sense complex geometrical structures (large curvature variations). The locations of these surfaces are initialized over the voxel space and are learned during training phase. We propose SplineNet, a deep network consisting of B-spline surfaces or classification of input 3D data represented in volumetric grid. We derive analytical solutions for updates of B-spline surfaces during back propagation. We show results on publicly available dataset and achieve superior performance as compared to state-of-the-art method. Full paper: pdf Centre for Visual Information Technology |
||||||||
Copyright © 2009 - IIIT Hyderabad. All Rights Reserved. |