RGB-D object discovery via multi-scene analysis
Authors:
- 1 September 2011
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 4850-4856
Abstract
We introduce an algorithm for object discovery from RGB-D (color plus depth) data, building on recent progress in using RGB-D cameras for 3-D reconstruction. A set of 3-D maps are built from multiple visits to the same scene. We introduce a multi-scene MRF model to detect objects that moved between visits, combining shape, visibility, and color cues. We measure similarities between candidate objects using both 2-D and 3-D matching, and apply spectral clustering to infer object clusters from noisy links. Our approach can robustly detect objects and their motion between scenes even when objects are textureless or have the same shape as other objects.Keywords
This publication has 16 references indexed in Scilit:
- Toward object discovery and modeling via 3-D scene comparison2011
- Discriminative clustering for image co-segmentation2010
- Object Recognition in 3D Point Clouds Using Web Data and Domain AdaptationThe International Journal of Robotics Research, 2010
- A Topic-Motion Model for Unsupervised Video Object Discovery2007
- Unsupervised Learning of Categories from Sets of Partially Matching Image Features2006
- Map building with mobile robots in dynamic environments2004
- Multiclass spectral clustering2003
- Fast approximate energy minimization via graph cutsIEEE Transactions on Pattern Analysis and Machine Intelligence, 2001
- Object modelling by registration of multiple range imagesImage and Vision Computing, 1992
- Random sample consensusCommunications of the ACM, 1981