Multi-Table Reinforcement Learning for Visual Object Recognition

  • Monica Piñol
  • Angel D. Sappa
  • Ricardo Toledo
Conference paper
Part of the Lecture Notes in Electrical Engineering book series (LNEE, volume 221)


This paper presents a bag of feature based method for visual object recognition. Our contribution is focussed on the selection of the best feature descriptor. It is implemented by using a novel multi-table reinforcement learning method that selects among five of classical descriptors (i.e., Spin, SIFT, SURF, C-SIFT and PHOW) the one that best describes each image. Experimental results and comparisons are provided showing the improvements achieved with the proposed approach.


Object recognition Artificial intelligence Reinforcement learning 



This work was partially supported by the Spanish Government under Research Program Consolider Ingenio 2010: MIPRCV (CSD2007-00018) and Project TIN2011-25606. Monica Piñol was supported by Universitat Autònoma de Barcelona grant PIF 471-01-8/09.


  1. 1.
    Csurka G, Dance CR, Fan L, Willamowski J, Bray C (2004) Visual categorization with bags of keypoints. In: Workshop on statistical learning in computer vision, Proceedings of the European conference on computer vision (2004), pp 1–22Google Scholar
  2. 2.
    Fei-Fei L, Perona P (2005) A bayesian hierarchical model for learning natural scene categories. In: Proceedings of IEEE conference on computer vision and pattern recognition, pp 524–531Google Scholar
  3. 3.
    Bay H, Tuytelaars T, Gool LV (2006) Surf: speeded up robust features. In: Proceedings of the European conference on computer vision, pp 404–417Google Scholar
  4. 4.
    Bosch A, Zisserman A, Muñoz X (2007) Image classification using random forests and ferns. In: Proceedings of international conference on computer visionGoogle Scholar
  5. 5.
    Piñol M, Sappa AD, López A, Toledo R (2012) Feature selection based on reinforcement learning for object recognition. In: adaptive learning agent workshop, pp 4–8Google Scholar
  6. 6.
    Shokri M, Tizhoosh HR (2008) A reinforcement agent for threshold fusion. Appl Soft Comput 8:174–181Google Scholar
  7. 7.
    Sahba F, Tizhoosh HR, Salama M (2007) Application of opposition-based reinforcement learning in image segmentation. In: IEEE Symposium on Computational Intelligence in Image and Signal Processing, Honolulu, HI, pp 246–251Google Scholar
  8. 8.
    Harandi MT, Ahmadabadi, MN, Araabi, BN (2004) Face recognition using reinforcement learning. Proc IEEE Int conf image process 4:2709–2712Google Scholar
  9. 9.
    Häming K, Peters G (2010) Learning scan paths for object recognition with relational reinforcement learning. In: Proceedings of the 7th IASTED international conference on signal processing, pattern recognition and applications, vol 678. Innsbruck, Austria, p 253Google Scholar
  10. 10.
    Jodogne S (2005) Reinforcement learning of perceptual classes using q learning updates. In: Proceedings of the 23rd IASTED international multi-conference on artificial intelligence and applications, pp 445–450Google Scholar
  11. 11.
    Jodogne S, Piater JH (2004) Interactive selection of visual features through reinforcement learning. In: Proceedings of 24th SGAI international conference on innovative techniques and applications of artificial intelligence, pp 285–298Google Scholar
  12. 12.
    Bianchi R, Ramisa A, de Mántaras R (2010) Automatic selection of object recognition methods using reinforcement learning. Adv Mach Learn 1:421–439Google Scholar
  13. 13.
    Sutton R, Barto A (1998) Reinforcement learning: an introduction Cambridge Univ Press, MA (1998)Google Scholar
  14. 14.
    Watkins CJCH (1989) Learning from delayed rewards. Ph.D. thesis, King’s College, CambridgeGoogle Scholar
  15. 15.
    Mitchell TM (1997) Machine learning. McGraw-Hill Science/Engineering/Math, New YorkGoogle Scholar
  16. 16.
    Leibe B, Schiele B (2003) Analyzing appearance and contour based methods for object categorization. Proc IEEE Conf Comput Vis Pattern Recogn 2:409Google Scholar
  17. 17.
    Ruzon MA, Tomasi C (2001) Edge, junction, and corner detection using color distributions. IEEE Trans Pattern Anal Mach Intell 23:1281–1295Google Scholar
  18. 18.
    Martin DR, Fowlkes CC, Malik J (2004) Learning to detect natural image boundaries using local brightness, color, and texture cues. IEEE Trans Pattern Anal Mach Intell 26:530–549Google Scholar
  19. 19.
    Mikolajczyk K, Schmid C (2005) A performance evaluation of local descriptors. IEEE Trans on Pattern Anal Mach Intell 27(10):1615–1630Google Scholar
  20. 20.
    Lowe D (2004) Distinctive image features from scale invariant keypoints. Int J Comput Vision 2:91–110CrossRefGoogle Scholar
  21. 21.
    van de Sande KEA, Gevers T, Snoek CGM (2010) Evaluating color descriptors for object and scene recognition. IEEE Trans Pattern Anal Mach Intell 32:1582–1596Google Scholar
  22. 22.
    Lazebnik S, Schmid C, Ponce J (2005) A sparse texture representation using local affine regions. IEEE Trans Pattern Anal Mach Intell 27(8):1265–1278Google Scholar
  23. 23.
    Nene SA, Nayar SK, Murase H (1996) Columbia object image library (COIL-100). Technical report (Feb 1996)Google Scholar

Copyright information

© Springer India 2013

Authors and Affiliations

  • Monica Piñol
    • 1
  • Angel D. Sappa
    • 1
  • Ricardo Toledo
    • 1
  1. 1.Computer Vision Center and Computer Science DepartmentUniversitat Autònoma de BarcelonaBarcelonaSpain

Personalised recommendations