Joint Color-Spatial-Directional clustering and Region Merging (JCSD-RM) for unsupervised RGB-D image segmentation

Abul Hasnat 1 Olivier Alata 1 Alain Treméau 1
1 LHC - Laboratoire Hubert Curien [Saint Etienne]
Abstract : Recent advances in depth imaging sensors provide easy access to the synchronized depth with color, called RGB-D image. In this paper, we propose an unsupervised method for indoor RGB-D image segmentation and analysis. We consider a statistical image generation model based on the color and geometry of the scene. Our method consists of a joint color-spatial-directional clustering method followed by a statistical planar region merging method. We evaluate our method on the NYU depth database and compare it with existing unsupervised RGB-D segmentation methods. Results show that, it is comparable with the state of the art methods and it needs less computation time. Moreover, it opens interesting perspectives to fuse color and geometry in an unsupervised manner.
Keywords : Unsupervised Clustering RGB-D image segmentation Directional distributions Bregman divergence Mixture model Region adjacency graph Region merging
Document type :
Journal articles
Complete list of metadatas

Cited literature [44 references]  Display  Hide  Download
https://hal-ujm.archives-ouvertes.fr/ujm-01273670
Contributor : Olivier Alata <>
Submitted on : Friday, February 12, 2016 - 5:32:26 PM
Last modification on : Thursday, July 26, 2018 - 1:10:43 AM
Long-term archiving on : Saturday, November 12, 2016 - 7:58:59 PM

File

pami_jcsd_rm_rev_draft_2.pdf
Files produced by the author(s)

Identifiers

Collections

UNIV-ST-ETIENNE | IOGS | PARISTECH | UNIV-PARIS-SACLAY

Citation

Abul Hasnat, Olivier Alata, Alain Treméau. Joint Color-Spatial-Directional clustering and Region Merging (JCSD-RM) for unsupervised RGB-D image segmentation. IEEE Transactions on Pattern Analysis and Machine Intelligence, Institute of Electrical and Electronics Engineers, 2016, ⟨10.1109/TPAMI.2015.2513407⟩. ⟨ujm-01273670⟩

Export

BibTeX TEI DC DCterms EndNote

Share

Metrics

Record views

157

Files downloads

230

Contact

support.ccsd.cnrs.fr
hal.support@ccsd.cnrs.fr
Logo CCSD