UNSUPERVISED FEATURE LEARNING FOR MID-LEVEL DATA REPRESENTATION

Yıl 2016, Cilt: 12 Sayı: 1, 51 - 79, 24.04.2016

Emrah Ergül Mehmet Karayel Oğuzhan Timuş Erkan Kıyak Turkish Naval Academy

Öz

Attribute based approaches are commonly used in recent
years instead of  low level features for
image classification which is one of the most important problems in the field
of computer vision. The most important advantage of attribute based approach is
that learning can be performed similar to human by using attributes which makes
sense for people. In this study, unsupervised attributes are developed in order
to avoid human related problems in supervised attribute learning. In our
proposed work, the attributes are generated as random binary and relative
definitions. The process of random attribute generation simplifies the data
modeling when compared to other work in the literature. In addition, a major
problem which is the increasing the numbers of attributes in attribute based
approaches is eliminated owing to the increasing the numbers of attributes
easily. Furthermore, attributes are selected more wisely using simple
applicable algorithm to improve the discriminative capacity of randomly
generated attribute set for image classification. The proposed approaches are
evaluated with the other similar attribute based studies comparatively in the
literature based on the same data set (OSR-Open Scene Recognition). Experiments
show that noteworthy performance increase is achieved.

Anahtar Kelimeler

Relative Attributes, Unsupervised Feature Extraction, Attribute Selection, Visual Recognition

Kaynakça

• Ferrari V. and Zisserman A. “Learning visual attributes” Advances in Neural Information Processing Systems, Vancouver CA, December 2007.
• Lampert C.H., Nickisch H. and Harmeling S. “Attribute-Based classification for zero-shot visual object categorization” IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 36, no. 3, 2014.
• Lampert C. H., Nickisch H., and Harmeling S. "Learning To Detect Unseen Object Classes by Between-Class Attribute Transfer" Proc. IEEE Conf. Computer Vision and Pattern Recognition (CVPR), 2009.
• Farhadi A., Endres I. and Hoiem D. “Attribute-centric recognition for cross- category generalization” CVPR, 2010.
• Farhadi A., Endres I., Hoiem D. and Forsyth D. “Describing objects by their attributes” CVPR 2009.
• Parikh D. and Grauman K. “Relative attributes” Int’l Conference on Computer Vision (ICCV), 2011.
• Sharma G., Jurie F., and Schmid C. “Expanded parts model for human attribute and action recognition in still images” CVPR, pp. 652 – 659, 2013.
• Akata Z., Perronnin F., Harchaoui Z. and Schmid C. “Label-embedding for attribute-based classification” CVPR, pp. 819 – 826, 2013.
• Tamara L.B., Alexander C.B. and Jonathan S. “Automatic attribute discovery and characterization from noisy web data” ECCV, pp. 663-676, [10] Russakovsky O. and Fei-Fei L. “Attribute learning in large-scale datasets” ECCV Workshops, pp. 1-14, 2010.
• Biswas A. and Parikh D. “Simultaneous active learning of classifiers & attributes via relative feedback” CVPR, 2013.
• Parkash A. and Parikh D. “Attributes for Classifier Feedback” European Conference on Computer Vision (ECCV), vol. 3, pp. 354-368, 2012.
• Rastegari M., Diba A., Parikh D., Farhadi A. “Multi-attribute queries: To merge or not to Merge” CVPR, 2013.
• Kumar N., Berg A.C., Belhumeur P. N., and Nayar S. K. “Attribute and smile classifiers for face verification” ICCV, 2009.
• Ma S., Sclaroff S. and Cinbis N.I. "Unsupervised learning of discriminative relative visual attributes" ECCV Workshop on Parts and Attributes, 2012.
• Karayel M. and Arica N. “Random attributes for image classification” IEEE 21th Conference on Signal Processing and Communications Applications, 2013.
• Wang Y. and Mori G. “A discriminative latent model of object classes and attributes” ECCV, pp. 155-168, 2010.
• Yu F.X., Ji R., Tsai M., Ye G. and Chang S. “Weak attributes for large- scale image retrieval” CVPR, 2012.
• Chen K., Gong S., Xiang T. and Loy C.C. “Cumulative attribute space for age and crowd density estimation” CVPR, pp. 2467 – 2474, 2013.
• Yu F.X., Cao L., Feris R.S., Smith J.R. and Chang S. “Designing category- level attributes for discriminative visual recognition” CVPR, 2013.
• Li W., Yu Q., Sawhney H. and Vasconcelos N. “Recognizing activities via bag of words for attribute dynamics” CVPR, pp. 2587 – 2594, 2013.
• Ma Z., Yang Y., Xu Z., Sebe N., Yan S. and Hauptmann A.G. “Complex event detection via multi-source video attributes” CVPR, 2013.
• Chen H., Gallagher A. and Girod B. “What's in a name: first names as facial attributes” CVPR, 2013.
• Sadovnik A., Gallagher A. and Chen T. "It's not polite to point: describing people with uncertain attributes" CVPR, 2013.
• Choi J., Rastegari M., Farhadi A. and Davis L.S. “Adding unlabeled samples to categories by learned attributes” CVPR, 2013.
• Wah C. and Belongie S. “Attribute-based detection of unfamiliar classes with humans in the loop” CVPR, pp. 779 – 786, 2013.
• Wang S., Joo J., Wang Y., and Zhu S.C. “Weakly supervised learning for attribute localization in outdoor scenes” CVPR, 2013.
• Saleh B., Farhadi A. and Elgammal A. “Object-centric anomaly detection by attribute-based reasoning,” CVPR, 2013.
• Bosch A., Xavier M. and Marti R. “A review: which is the best way to organize/classify images by content?” Image and Vision Computing, 2006.
• Ergül E., Ertürk S. and Arica N. “Unsupervised Relative Attribute Extraction” IEEE 21th Conference on Signal Processing and Communications Applications, 2013.
• Chang C.C. and Lin C.J. “LIBSVM : A library for support vector machines” ACM Transactions on Intelligent Systems and Technology, pp. 1-27, [32] Shrivastava A., Singh S. and Gupta A. "Constrained semi-supervised learning using attributes and comparative attributes", ECCV, vol 3, pp. 369- 383. 2012.
• Yu, A., and Grauman, ,K., “Just Noticeable Differences in Visual Attributes” ICCV, 2015.
• Verma, Y., and Jawahar, C.V., “Exploring Locally Rigid Discriminative Patches for Learning Relative Attributes” ICCV, 2015.
• Alpaydın E., “Support Vector Machines,” in Introduction to machine Learning, The MIT Press, London, 2004, pp. 218-225. [36]
• Learning, vol. 20, no. 3, pp. 273-297, 1995. [37]
• (Weka) Versiyon 3.7.11, Waikato University, Hamilton, 2014.
• Coates A., Lee H. and Andrew Y. Ng. “An analysis of single-layer networks in unsupervised feature Learning,” International Conference on Artificial Intelligence and Statistics (AISTATS), 2011.