Adaptive 3-D Object Classification with Reinforcement Learning

Jens Garstka, Gabriele Peters

2015

Abstract

We propose an adaptive approach to 3-D object classification. In this approach appropriate 3-D feature descriptor algorithms for 3-D point clouds are selected via reinforcement learning depending on properties of the objects to be classified. This approach is supposed to be able to learn strategies for an advantageous selection of 3-D point cloud descriptor algorithms in an autonomous and adaptive way, and thus is supposed to yield higher object classification rates in unfamiliar environments than any of the single algorithms alone. In addition, we expect our approach to be able to adapt to subsequently added 3-D feature descriptor algorithms as well as to autonomously learn new object categories when encountered in the environment without further user assistance. We describe the 3-D object classification pipeline based on local 3-D features and its integration into the reinforcement learning environment.

References

  1. Alexandre, L. A. (2012). 3D descriptors for object and category recognition: a comparative evaluation. In Workshop on Color-Depth Camera Fusion in Robotics at the IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), Vilamoura, Portugal.
  2. Cholewa, M. and Sporysz, P. (2014). Classification of dynamic sequences of 3d point clouds. In Artificial Intelligence and Soft Computing, pages 672-683. Springer.
  3. Filipe, S. and Alexandre, L. A. (2013). A Comparative Evaluation of 3D Keypoint Detectors. In 9th Conference on Telecommunications, Conftele 2013, pages 145-148, Castelo Branco, Portugal.
  4. Frome, A., Huber, D., Kolluri, R., Bulow, T., and Malik, J. (2004). Recognizing objects in range data using regional point descriptors. In Proceedings of the European Conference on Computer Vision (ECCV).
  5. Guo, Y., Bennamoun, M., Sohel, F., Lu, M., and Wan, J. (2014). 3d object recognition in cluttered scenes with local surface features: A survey. IEEE Transactions on Pattern Analysis and Machine Intelligence, 36(11).
  6. Johnson, A. E. and Hebert, M. (1998). Surface matching for object recognition in complex three-dimensional scenes. Image and Vision Computing, 16(9):635-651.
  7. Lai, K., Bo, L., Ren, X., and Fox, D. (2011). A largescale hierarchical multi-view rgb-d object dataset. In Robotics and Automation (ICRA), 2011 IEEE International Conference on, pages 1817-1824. IEEE.
  8. Rusu, R., Blodow, N., and Beetz, M. (2009). Fast point feature histograms (fpfh) for 3d registration. In Robotics and Automation, 2009. ICRA 7809. IEEE International Conference on, pages 3212-3217.
  9. Rusu, R. B., Blodow, N., Marton, Z. C., and Beetz, M. (2008). Aligning point cloud views using persistent feature histograms. In Intelligent Robots and Systems, 2008. IROS 2008. IEEE/RSJ International Conference on, pages 3384-3391. IEEE.
  10. Salti, S., Tombari, F., and Stefano, L. D. (2011). A performance evaluation of 3d keypoint detectors. In 3D Imaging, Modeling, Processing, Visualization and Transmission (3DIMPVT), 2011 International Conference on, pages 236-243. IEEE.
  11. Stein, F. and Medioni, G. (1992). Structural indexing: efficient 3-d object recognition. IEEE Trans. PAM, 14:125-145.
  12. Sutton, R. S. and Barto, A. G. (1998). Reinforcement learning: An introduction, volume 1. Cambridge Univ Press.
  13. Toldo, R., Castellani, U., and Fusiello, A. (2010). The bag of words approach for retrieval and categorization of 3d objects. The Visual Computer, 26(10):1257-1268.
  14. Tombari, F., Salti, S., and Di Stefano, L. (2010a). Unique shape context for 3d data description. In Proceedings of the ACM workshop on 3D object retrieval, pages 57-62. ACM.
  15. Tombari, F., Salti, S., and Di Stefano, L. (2010b). Unique signatures of histograms for local surface description. In Computer Vision-ECCV 2010, pages 356-369. Springer.
  16. Watkins, C. and Dayan, P. (1992). Technical note: Qlearning. Machine Learning, 8(3-4):279-292.
  17. Watkins, C. J. C. H. (1989). Learning from Delayed Rewards. PhD thesis, King's College, Cambridge, UK.
  18. Wiering, M. and Van Otterlo, M. (2012). Reinforcement learning. In Adaptation, Learning, and Optimization, volume 12. Springer.
  19. Wu, C.-C. and Lin, S.-F. (2011). Efficient model detection in point cloud data based on bag of words classification. Journal of Computational Information Systems, 7(12):4170-4177.
  20. Zhong, Y. (2009). Intrinsic shape signatures: A shape descriptor for 3d object recognition. In Computer Vision Workshops (ICCV Workshops), 2009 IEEE 12th International Conference on, pages 689-696. IEEE.
Download


Paper Citation


in Harvard Style

Garstka J. and Peters G. (2015). Adaptive 3-D Object Classification with Reinforcement Learning . In Proceedings of the 12th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO, ISBN 978-989-758-123-6, pages 381-385. DOI: 10.5220/0005563803810385


in Bibtex Style

@conference{icinco15,
author={Jens Garstka and Gabriele Peters},
title={Adaptive 3-D Object Classification with Reinforcement Learning},
booktitle={Proceedings of the 12th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO,},
year={2015},
pages={381-385},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0005563803810385},
isbn={978-989-758-123-6},
}


in EndNote Style

TY - CONF
JO - Proceedings of the 12th International Conference on Informatics in Control, Automation and Robotics - Volume 2: ICINCO,
TI - Adaptive 3-D Object Classification with Reinforcement Learning
SN - 978-989-758-123-6
AU - Garstka J.
AU - Peters G.
PY - 2015
SP - 381
EP - 385
DO - 10.5220/0005563803810385