Unsupervised Object Category Recognition from Image Datasets | Saima | Nova Journal of Engineering and Applied Sciences

Unsupervised Object Category Recognition from Image Datasets

Sultana Saima, Rahman Syeda Ohishee, Hossain Alamgir


The variability of the object models is treated as flexible constellations of rigid parts which is represented by a joint probability density function (pdf) on the shape of the constellation and the output of part detectors. Firstly we use method an Affine Invariant Salient Region Detector to identify the distinctive parts in the training set for clustering data. Then we use Markov chain Monte Carlo expectation maximization (MCMC-EM) algorithm to learn the statistical shape model of the object and discover object categories in an unsupervised manner. In the MCMC-EM algorithm, the high-dimensional integrals required in the EM algorithm are estimated using MCMC sampling. The MCMC sampler requires simulation of sample paths from a continuous time Markov process, conditional on the beginning and ending states and the paths of the neighboring sites.

Keywords: Object recognition, dataset, algorithm, models, detector, probability density function (pdf).

Full Text:




Russell BC, Torralba A, Murphy KP, Freeman WT. LabelMe: a database and web-based tool for image annotation. IJCV. 2008; 77(1–3):157–173.

Winn J, Criminisi A, Minka T. Object categorization by learned universal visualdictionary. In ICCV. 2005; I:756-763.

Viola P, Jones MJ. Robust real-time face detection. Int. J. of Comp. Vision. 2004; 57(2):137–154.

Leibe B, Seemann E, Schiele B. Pedestrian detection in crowded scenes. In Proceedings of the IEEE Conference on Computer Vision and PatternRecognition, 2005.

Mikolajczyk K, Schmid C, Zisserman A. Human detection based on a probabilistic assembly of robust part detectors. In Proceedings of theEuropean Conference on Computer Vision, 2004.

Grauman K, Darrell T. Unsupervised learning of categories from sets of partiallymatching image features. In CVPR, 2006.

Sivic J, Russell BC, Efros AA, Zisserman A, Freeman WT. Discoveringobjects and their location in images. In ICCV, 2005; 370–377.

Russell B, Efros AA, Sivic J, Freeman WT, Zisserman A. Using multiple segmentations to discover objects and their extent in image collections. In CVPR, 2006.

Todorovic S, Ahuja N. Extracting subimages of an unknown category from a set of images. In CVPR, 2006.

Carbonetto P, Dork´o G, Schmid C, Kuck H, de Freitas H. Learning to recognize objects with little supervision. International Journal of Computer Vision, 2008; 77 : 219-237.

Edelman S, Intrator N. Unsupervised statistical learning in vision:computational principles, biological evidence. Miscellaneous Papers, Available from kybele.psych.cornell.edu, 23 April, 2004.

Kadir T, Zisserman A, Brady M. An affine invariant salient regiondetector. Proceedings of the 8th European Conference on Computer Vision, Prague, Czech Republic. 2004.

Kadir T, Brady JM. Scale, saliency and image description. Intl. J. of Computer Vision, 2001; 5(2):83-105.

Kamarainen JK, Hamouz M, Kittler J, Paalanen P, Ilonen J, Drobchenko A. ObjectLocalisation Using Generative Probability Model for Spatial Constellation and Local Image Features. ICCV, pp.1-8, 2007 and IEEE 11th International Conference on Computer Vision, 2007.

Sahbi H, Audibert J, Keriven R. Context-Dependent Kernels for Object Classification. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2011; 33(4) : 699- 708.

Wu W, Yang J. Semi-Supervised Learning of Object Categories from Paired Local Features. ACM International Conference on Image and Video Retrieval, CIVR 2008, Niagra Falls, Canada, 07. July 2008.

Halder A, Pramanik S, Kar A. Dynamic Image Segmentation using Fuzzy C-Means based Genetic Algorithm. International Journal of Computer Applications. 2011; 28(6) : (0975- 8887.

Halder A, Pramanik S. An Unsupervised Dynamic Image Segmentationusing Fuzzy Hopfield Neural Network based Genetic Algorithm. International Journal of Computer Science Issues. 2012; 9(2).

Cho M, Lee J, Lee KM. Feature correspondence and deformable object matching via agglomerative correspondence clustering. Computer vision, 2009 IEEE 12th International Conference on, date of conference: Sept. 29 2009- Oct. 2 2009: 1280-1287.

Felzenszwalb PF, Girshick, McAllester. Discriminatively Trained Deformable Part Models. This project has been supported by the National Science Foundation under grant no. 0534820, 0746569 and 0811340. http://people.cs.uchicago.edu/~pff/latent-release4/.

Kokkinos I. the Project-Teams GALEN.Rapid Deformable Object Detection using Bounding-based Techniques. Informatique/Vision par ordinateur et reconnaissance de formes, RR-7940, ISSN 0249-6399 ISRN INRIA/RR--7940-FR+ENG, research Report no 7940. 2012: 19.

Kashem MA, Akhter MN, Ahmed S, MahbubAlam M. Face Recognition System Based on Principal Component Analysis (PCA) withBack Propagation Neural Networks (BPNN). Canadian Journal on Image Processing and Computer Vision. 2011; 2 (4).

Do T, Kijak E. Face Recognition Using Co-occurrence Histograms of Oiented Gradients. IEEE International Congerence on Acoustics, Speech and Signal Processing, Kyoto International Conference Center, Koyoto, Japan, paper: IVMSP-P9.2,session: Feature Extraction and Analysis, location: poster Area E, presentation time, March 29, 2012, presentation: poster.

Santhosh S, Manjula VS. Face Emotion Analysis Using Gabor FeaturesIn Image Database for Crime Investigation. International Journal of Data Engineering. 2011; 2 (2).

Arumugam D, Purushothaman S. Emotion Classification Using Facial Expression. International Journal of Advanced Computer Science andApplications. 2011; 2(7).

DOI: http://dx.doi.org/10.20286/jeas.v3i3.24


  • There are currently no refbacks.

Nova Journal of Engineering and Applied Sciences ISSN: 2292-7921

Nova Pub inc.

Nova Explore Publications is a member of CrossRef.
DOI Prefix: 10.20286