Unsupervised feature learning for 3D scene labeling
Authors:
Top Cited Papers
- 1 May 2014
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- No. 10504729,p. 3050-3057
Abstract
This paper presents an approach for labeling objects in 3D scenes. We introduce HMP3D, a hierarchical sparse coding technique for learning features from 3D point cloud data. HMP3D classifiers are trained using a synthetic dataset of virtual scenes generated using CAD models from an online database. Our scene labeling system combines features learned from raw RGB-D images and 3D point clouds directly, without any hand-designed features, to assign an object label to every 3D point in the scene. Experiments on the RGB-D Scenes Dataset v.2 demonstrate that the proposed approach can be used to label indoor scenes containing both small tabletop objects and large furniture pieces.Keywords
This publication has 18 references indexed in Scilit:
- Multipath Sparse Coding Using Hierarchical Matching Pursuit2013
- Histograms of Sparse Codes for Object Detection2013
- Real-time human pose recognition in parts from single depth imagesCommunications of the ACM, 2013
- Detection-based object labeling in 3D scenes2012
- A learned feature descriptor for object recognition in RGB-D data2012
- Going into depth: Evaluating 2D and 3D cues for object classification on a new, large-scale object dataset2011
- KinectFusion: Real-time dense surface mapping and tracking2011
- Scene parsing using a prior world modelThe International Journal of Robotics Research, 2011
- Convolutional deep belief networks for scalable unsupervised learning of hierarchical representations2009
- The princeton shape benchmark2004