Publications

Our research interests include machine learning and computer vision. In machine learning, we are interested in pattern classification (especially large margin and kernel classifiers), clustering, kernel methods, data modeling, statistical techniques, manifold learning and distance metric learning. In computer vision, we are particularly interested in visual object recognition, object category localization, face detection, face recognition and image retrieval. Our goal is to combine image analysis tools with the machine learning and statistical modeling techniques.

Candidates who are willing to study on one of the research topics listed above are welcome to contact us.

Publications in 2020

Semi-Supervised Robust Deep Neural Networks for Multi-Label Image Classification

Hakan Cevikalp, Burak Benligiray, Omer Nezih Gerek

Pattern Recognition.

Publications in 2019

Face recognition based on videos by using convex hulls

Hakan Cevikalp, Hasan Serhan Yavuz, Bill Triggs

IEEE Transactions on Circuits and Systems for Video Technology, 2019.

High-dimensional data clustering by using local affine/convex hulls

Hakan Cevikalp

Pattern Recognition Letters, 128, 2019, pp.427-432.

Discriminatively Learned Convex Models for Set Based Face Recognition

Hakan Cevikalp, Golara Ghorban Dordinejad

IEEE International Conference on Computer Vision, 2019.

Detection Based Tracking of Unmanned Aerial Vehicles

Bedirhan Uzun, Onur Eker, Hasan Saribas, Hakan Cevikalp

2019 27th Signal Processing and Communications Applications Conference (SIU)

Semi-Supervised Robust Deep Neural Networks for Multi-Label Classification

Hakan Cevikalp, Burak Benligiray, Omer Nezih Gerek, Hasan Saribas

IEEE Conference on Computer Vision and Pattern Recognition Workshops. 2019.

Visual Object Tracking by Using Ranking Loss

Hakan Cevikalp, Hasan Saribas, Burak Benligiray, Sinem Kahvecioglu

IEEE International Conference on Computer Vision Workshops, 2019.

Polyhedral Conic Classifiers for Computer Vision Applications and Open Set Recognition

Hakan Cevikalp, Halil Saglamlar

IEEE transactions on pattern analysis and machine intelligence, 2019

A Hybrid Method for Tracking of Objects by UAVs

Hasan Saribas, Bedirhan Uzun, Burak Benligiray, Onur Eker, Hakan Cevikalp

IEEE Conference on Computer Vision and Pattern Recognition Workshops, 2019.

Visual Object Detection System for Autonomous Vehicles in Smart Factories

Nazlıcan Gengeç, Hakan Çevikalp, Hasan Serhan Yavuz, Ahmet Yazıcı

Innovations in Intelligent Systems and Applications Conference (ASYU)

Publications in 2018

Large-scale image retrieval using transductive support vector machines

Hakan Cevikalp, Merve Elmas, Savas Ozkan

Computer Vision and Image Understanding, 2018.

Recurrent Neural Networks for Remote Sensing Image Classification

Mohamed Lakhal, Hakan Cevikalp, Sergio Escalera, Ferda Ofli

IET Computer Vision, 2018.

CRN: End-to-end convolutional recurrent neural network structure applied to vehicle classification

Mohamed Lakhal, Sergio Escalera, Hakan Cevikalp

International Conference on Computer Vision Theory and Applications (VISAPP), 2018.

Car Localization in Aerial Images Taken from Quadcopter

Hasan Saribas, Hakan Cevikalp, Sinem Kahvecioglu

International Conference on Recent Trends in Engineering and Technology (RTET--2018), 2018.

Publications in 2017

Best Fitting Hyperplanes for Classification

Hakan Cevikalp

IEEE Transactions on Pattern Analysis and and Machine Intelligence, vol 39, pp. 1076-1088, 2017.

Visual Object Detection Using Cascades of Binary and One-Class Classifiers

Hakan Cevikalp, Bill Triggs

International Journal of Computer Vision, vol 123, pp. 123-334, 2017.

Fast and Accurate Face Recognition with Image Sets

Hakan Cevikalp, Hasan S. Yavuz

International Conference on computer Vision (ICCV) Workshops, 2017.

Large-scale Robust Transductive Support Vector Machines

Hakan Cevikalp, Vojtech Franc

Neurocomputing, vol. 235, pp.199-209, 2017.

The Fourier Transform Based Descriptor for Visual Object Classification

Hakan Cevikalp, Zuhal Kurt

Anadolu University Journal of Science and Technology A- Applied Sciences and Engineering, vol. 18, pp.247-261, 2017.

Polyhedral Conic Classifiers for Visual Object Detection and Classification

Hakan Cevikalp, Bill Triggs

IEEE Society Conference on Computer Vision and Pattern Recognition (CVPR), 2017.

Publications in 2016

Towards Category Based Large-Scale Image Retrieval Using Transductive Support Vector Machines

Hakan Cevikalp, Merve Elmas, Savas Ozkan

Europeran Conference on Computer Vision Workshops, 2016.

Publications in 2015

A Comprehensive Comparison of Features and Embedding Methods for Face Recognition

Hasan S. Yavuz, Hakan Cevikalp, Rifat Edizkan

Turkish Journal of Electrical Engineering & Computer Sciences, 2015.

Towards Large-Scale Face Recognition Based on Videos

Meltem Yalcin, Hakan Cevikalp, Hasan S. Yavuz

International Conference on Computer Vision Workshop on Video Summarization for Large-Scale Analytics, 2015.

Publications in 2014

Semi-supervised discriminative common vector method for computer vision applications

Hakan Cevikalp

Neurocomputing, vol. 129, pp. 289-297, 2014.

2-Sided Best Fitting Hyperplane Classifier

Hakan Cevikalp

International Conference on Pattern Recognition, 2014.

Semi-Automatic Annotation Tool

Tugce Dongel, Hakan Cevikalp

IEEE Signal Processing and Communications Applications Conference (in Turkish), 2014.

Publications in 2013

Hyperdisk Based Large Margin Classifier

Hakan Cevikalp, Bill Triggs

Pattern Recognition, vol. 46, pp. 1523-1531, 2013.

Face and Landmark Detection by Using Cascade of Classifier

Hakan Cevikalp, Bill Triggs, Vojtech

IEEE International Conference on Automatic Face and Gesture Recognition, 2013.

Return of the King: The Fourier Transform Based Descriptor for Visual Object Classification

Hakan Cevikalp,Zuhal Kurt, Ahmet O. Onarcan

IEEE Signal Processing and Communications Applications Conference, 2013.

Automatic face recognition from frontal images

asan S. Yavuz, Hakan Cevikalp, Rifat Edizkan

IEEE Signal Processing and Communications Applications Conference (in Turkish), 2013.

Publications in 2012

Distance Metric Learning by Quadratic Programming based on Equivalence Constraints

Hakan Cevikalp

International Journal of Innovative Computing, Information and Control, 2012.

Dijital Imgelerdeki Nesnelerin Konumlarini Bulmak Icin Yeni Bir Yontem

O. Akyuz, H. Cevikalp, and G. Usanmaz

IEEE Signal Processing and Communications Applications Conference, 2012.

Efficient Object Detection Using Cascades of Nearest Convex Model Classifiers

H. Cevikalp, B. Triggs

IEEE Society Conference on Computer Vision and Pattern Recognition (CVPR), 2012.

Non-speech Environmental Sound Classification Using SVMs with a New Set of Features

O. Akyuz, H. Cevikalp, and G. Usanmaz

International Journal of Innovative Computing, Information and Control, 2012.

Publications in 2011

Large Margin Classifier Based on Hyperdisks (nominated for the best paper award)

Hakan Cevikalp

International Conference on Machine Learning and Applicaitons, Honolulu, 2011.

Digital Comparisons of features for automatic eye and mouth localization

H.Cevikalp, H. S. Yavuz, R. Edizkan, H. Gunduz, C. M. Kandemir

INISTA, Istanbul, 2011.

Publications in 2010

Large margin classifiers based on affine hulls

H.Cevikalp, B. Triggs, H. S. Yavuz, Y. Kucuk, M. Kucuk, A. Barkana

Neurocomputing, 2010, Volume 73, 3160-3168.

Semi-supervised Distance Metric Learning in High-Dimensional Spaces by Using Equivalence Constraints

H.Cevikalp, H. S. Yavuz, R. Edizkan, H. Gunduz, C. M. Kandemir

Communications in Computer and Information Science, 2010, Volume 68, Part 4, 242-254.

Face Recognition Based on Image Sets

H.Cevikalp, B.Triggs

IEEE Conference on Computer Vision and Pattern Recognition, San Francisco USA 2010.

Semi-Supervised Distance Metric Learning by Quadratic Programming

Hakan Cevikalp

International Conference on Pattern Recognition, Istanbul Turkey 2010.

Large Margin Classifier Based on Affine Hulls

H.Cevikalp and H. S. Yavuz

International Conference on Pattern Recognition, Istanbul Turkey 2010.

Imge Setleriyle Yuz Tanima,(in Turkish)

Hakan Cevikalp

IEEE Signal Processing and Communications Applications Conference, Diyarbakir, 2010.

Publications in 2009

Large Margin Classifiers Based on Convex Class Models

H.Cevikalp and B.Triggs

2nd IEEE International Workshop on Subspace Methods (in conjunction with IEEE International Conference on Computer Vision), Kyoto, Japan, 2009

Multilinear-CLAFIC methods for tensor data in image recognition,(in Turkish)

H.Cevikalp and H. S. Yavuz

IEEE Signal Processing and Communications Applications Conference, Side, April 2009.

Semi-supervised distance metric learning for visual object classification

H. Cevikalp and R. Paredes

4th International Conference on Computer Vision Theory and Applications, Lisbon, Portugal, 2009.

Two-dimensional subspace classifiers for face recognition

H.Cevikalp and H. S. Yavuz

Neurocomputing,vol. 72, pp.1111-1120, 2009.

Publications in 2008

Comparison of Principal Component Analysis (PCA) and Multilinear Principal Component Analysis (MPCA) Methods for Face Recognition

H. S. Yavuz, M. A. Cay

International Conference on Multivariate Statistical Modelling & High Dimensional Data Mining (HDM-2008), Erciyes University, June 2008.

New distance measure for hierarchical clustering

H. S. Yavuz and H. Cevikalp

16th IEEE Signal Processing and Communications Applications Conference, Didim - Turkey, 2008.

Local classifier weighting by quadratic programming

H. Cevikalp and R. Polikar

IEEE Transactions on Neural Networks, vol. 19, pp. 1832-1838, 2008.

Nearest hyperdisk methods for high-dimensional classification

H. Cevikalp, B. Triggs, and R. Polikar

International Conference on Machine Learning, Helsinki, Finland, July 5-9, 2008.

Margin-based discriminant dimensionality reduction for visual recognition

H. Cevikalp, B. Triggs, F. Jurie, and R. Polikar

IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR), Anchorage, Alaska USA, June 24-26, 2008.

Semi-supervised dimensionality reduction using pairwise equivalence constraints

H. Cevikalp, J. Verbeek, F. Jurie, and A. Klaser

3rd International Conference on Computer Vision Theory and Applications, Funchal, Madeira Portugal, January 2008.

Manifold based local classifiers: linear and nonlinear approaches

H. Cevikalp, D. Larlus, M. Neamtu, B. Triggs, and F. Jurie

Journal of Signal Processing Systems, (online publishment),2008.

Publications in 2007

The Kernel Common Vector Method: A Novel Nonlinear Subspace Classifier for Pattern Recognition

H. Cevikalp, M. Neamtu, and A. Barkana

IEEE Trans. On Systems, Man, and Cybernetics Part B: Cybernetics, 37(3), pp. 937-951, 2007.

Local subspace classifiers: linear and nonlinear approaches

H. Cevikalp, D. Larlus, M. Douze, and F. Jurie

The 2007 IEEE Machine Learning for Signal Processing Workshop, August 2007, Thessaloniki, Greece.

A supervised clustering algorithm for the initialization of RBF Neural Network Classifiers

H. Cevikalp, D. Larlus, and F. Jurie

The 15th IEEE Signal Processing and Communications Applications Conference, Eskisehir, Turkey, June 2007.

Publications in 2006

Two Dimensional (2D) Subspace Classifiers for Image Recognition

Cevikalp, H. S. Yavuz, A. Barkana

14th European Signal Processing Conference EUSIPCO2006, Florence Italy, Sept. 2006.

Goruntu Tanimada Iki Boyutlu CLAFIC Yontemleri

H. S. Yavuz, H. Çevikalp, A. Barkana

The 14th IEEE Signal Processing and Communications Applications Conference, Antalya, Turkey, April 2006.

Discriminative common vector method with kernels

Cevikalp, M. Neamtu, and M. Wilkes

IEEE Trans. on Neural Networks, 17(5), pp.1550-1565, November 2006.

Theoretical analysis of linear discriminant analysis criteria

Hakan Cevikalp

The 14th IEEE Signal Processing and Communications Applications Conference, Antalya, Turkey, April 2006.

Two dimensional (2D) subspace classifiers for image recognition

H. Cevikalp, H. S. Yavuz, A. Barkana

The 14th European Signal Processing Conference, Florence, Italy, September 2006.

Publications in 2005

Feature Extraction Tehniques in High-Dimensional Spaces: Linear and Nonlinear Approaches

Hakan Cevikalp

Ph. D. thesis, Vanderbilt University, 2005.

Discriminative common vectors for face recognition

H. Cevikalp, M. Neamtu, M. Wilkes, and A. Barkana

IEEE Trans. on Pattern Analysis and Machine Intelligence, Vol. 27, January 2005.

Nonlinear common vectors for pattern classification

H. Cevikalp and M. Neamtu

The 13th European Signal Processing Conference, Antalya, Turkey, September 2005.

A comparison of the common vector and the discriminative common vector methods for face recognition

H. Cevikalp, B. Barkana, and A. Barkana

The 9th World Multi-Conference on Systemics, Cybernetics and Informatics, July 2005, Orlando, USA.

Nonlinear common vectors for pattern classification

H. Cevikalp, M. Neamtu, and M. Wilkes

The 9th World Multi-Conference on Systemics, Cybernetics and Informatics, July 2005, Orlando, USA.

Nonlinear discriminant common vectors (Turkish)

H. Cevikalp and M. Neamtu

The 13th IEEE Signal Processing and Communications Applications Conference (SIU), Kayseri, Turkey, pp. 292-295, May 2005.

Publications in 2004

Phantom Validation of Coregistration of PET and CT for Image-guided Radiotherapy

W.C. Lavely, C. Scarfone, H. Cevikalp, R. Li, D. W. Byrne, A. J. Cmelak, B. M. Dawant, R. R. Price, D. E. Hallahan, J. M. Fitzpatrick

Medical Physics, 2004 May; 31(5):1083-92.

Face recognition by using discriminative common vectors

H. Cevikalp, and M. Wilkes

17th Int. Conf. on Pattern Recognition (ICPR), Cambridge, UK, pp.326-329, August 2004.

A novel method for face recognition (Turkish)

H. Cevikalp, M. Neamtu, M. Wilkes, and A. Barkana

The 12th IEEE Signal Processing and Communications Applications Conference (SIU), Cesme, Turkey, April 2004.

A new point of view to common vector approach (Turkish)

M. B. Gulmezoglu, H. Cevikalp and A. Barkana

The 12th IEEE Signal Processing and Communications Applications Conference (SIU), Cesme, Turkey, April 2004.

Publications in 2003

Phantom Validation of Coregistration of PET and CT for Image-guided Radiotherapy

W.C. Lavely, C. Scarfone, H. Cevikalp, R. Li, D. W. Byrne, A. J. Cmelak, R. R. Price, D. E. Hallahan, J. M. Fitzpatrick

Ann. Mtg. Radiol. Soc. N. America, Chicago, IL (Nov 2003).

Marker orientation in fiducial registration

X. Liu, H. Cevikalp, J. M. Fitzpatrick

Proc. SPIE Medical Imaging 2003, San Diego, CA 1176-1183 (Feb 2003).

Image-guided otologic surgery

R. Labadie, M. Fenlon, H. Cevikalp, S. Harris, R. Galloway, J.M. Fitzpatrick

Proc. 17th Int. Congress and Exhibition, Computer Assisted Radiology and Surgery, London, H.U. Lemke, et al, ed., Elsevier, Int. Congress Series 1256, 613-618 (June 2003).

Publications in 2002

Gurultulu verilerin taninmasinin iyilestirilmesi (Improving recognition rates for noisy signals)

H. Cevikalp, M. B. Gulmezoglu, and A. Barkana

The 11th Conference on Signal Processing and Applications, June 2002, Denizli.

Publications in 2001

The Comparison of Fuzzy Inference Systems and Neural Network Approaches with ANFIS Method for Fuel Consumption Data

H. Atmaca, B. Cetisli, H. S. Yavuz

Second International Conference on Electrical and Electronics Engineering ELECO’2001, Bursa-Turkey, Nov. 2001.

Publications in 2000

LPG Isiticilarin Bulanik Mantik ile Kontrolu

H. Atmaca, B. Cetisli, H. S. Yavuz

International Conference on Electrical and Electronics Engineering ELECO’2000, Bursa-Turkey, Nov. 2000.

Publications in 1999

The Implementation of Fuzzy Flip-Flops as Memory Modules

H. Atmaca, H. S. Yavuz

International Conference on Electrical and Electronics Engineering ELECO’99, Bursa-Turkey, Dec. 1999.