The broad goal of my research is to explore better ways to understand, interpret and manipulate visual data*. My research interests span a diverse set of topics, ranging from image editing to visual data mining, and to multimodal learning of language and vision.

Aykut Erdem is an Assistant Professor in the Department of Computer Engineering at Hacettepe University, where he is a member of the Computer Vision Lab. His research interests are in computer vision and machine learning, with much of his recent activities focused on image editing, video and image collection summarization, and combining language and vision. Prior to joining Hacettepe University in 2010, he worked as a post-doctoral researcher in the Computer Science Department of Ca’Foscari University of Venice in Italy. He received his PhD degree in Computer Engineering from METU in 2008, and before that, he obtained his BSc and MSc degrees from the same institution in 2001 and 2003, respectively. During his PhD studies, he was a visiting researcher at Virginia Tech during Summer 2004 and a visiting scholar at MIT for Fall 2007.

Highlights and News
June 2017 Our work on sampling-based image and video matting will be published in IEEE Transactions on Image Processing: "Alpha Matting with KL-Divergence Based Sparse Sampling".
May 2017 Slides from our "Adversarial Training and Generative Adversarial Networks" tutorial at SIU 2017 are now available online.
Dec 2016 Our work on evaluation of automatic metrics for image captioning is accepted to EACL 2017 as a long paper: "Re-evaluating Automatic Metrics for Image Captioning". slides
Dec 2016 Our paper on using GANs to generate outdoor scenes from attributes and semantic layouts is out on arXiv.
Nov 2016 Our work on learning dynamic saliency will be published in Signal Processing: Image Communication: "A Comparative Study for Feature Integration Strategies in Dynamic Saliency Estimation".
Jul 2016 Our paper on deep models for dynamic saliency estimation is out on arXiv.
May 2016 Slides from our "Deep Learning in Computer Vision" tutorial at SIU 2016 are now available online.
May 2016 Our paper on Turkish image captioning won the Alper Atalay Best Student Paper Award (First Prize) at SIU 2016.
May 2016 Our paper on photo collection summarization won the IEEE Best Student Paper Award (Second Prize) at SIU 2016.
Apr 2016 Our part-based tracking work will be published in Journal of Visual Communication and Image Representation: "Deformable Part-based Tracking by Coupled Global and Local Correlation Filters".
Apr 2016 Our work on Turkish Image Description Generation is featured on national TV.
Mar 2016 I am awarded a hardware donation (a Tesla K40 GPU) from NVIDIA for my research.
Feb 2016 A full paper accepted to Eurographics 2016: "An Objective Deghosting Quality Metric for HDR Images".
Jan 2016 I'll be speaking at YAZGIG 2016.
Jan 2016 We'll be giving a tutorial on deep learning at SIU 2016.
older items

Funding
Principal Investigator TUBITAK 3501 Career Development Program Award #113E497, 2014-2017 · project page
Co-Investigator TUBITAK 1007 Public Institutions Research Funding Program Award #114G028, 2016-2019 · project page
TUBITAK 1001 The Support Program for Scientific and Technological Research Projects Award #113E116 and European Union under European Cooperation in Science and Technology (COST) Programme: ICT COST IC1037 Action, 2014-2017 · project page
TUBITAK 3501 Career Development Program: Award #112E146, 2012-2015 · project page
Donations NVIDIA Hardware Donation Grant (Tesla K40 GPU), 2016

Selected Publications

Re-evaluating Automatic Metrics for Image Captioning

Re-evaluating Automatic Metrics for Image Captioning
Mert Kilickaya, Aykut Erdem, Nazli Ikizler-Cinbis, and Erkut Erdem
EACL 2017
pdf · slides

Learning to Generate Images of Outdoor Scenes from Attributes and Semantic Layouts

Learning to Generate Images of Outdoor Scenes from Attributes and Semantic Layouts
Levent Karacan, Zeynep Akata, Aykut Erdem, Erkut Erdem
arXiv preprint arXiv:1612.00215
pdf

Two-Stream Convolutional Networks for Dynamic Saliency Prediction

Two-Stream Convolutional Networks for Dynamic Saliency Prediction
Cagdas Bak, Aykut Erdem, Erkut Erdem
arXiv preprint arXiv:1607.04730
pdf

Deformable Part-based Tracking by Coupled Global and Local Correlation Filters

Deformable Part-based Tracking by Coupled Global and Local Correlation Filters
Osman Akin, Erkut Erdem, Aykut Erdem, Krystian Mikolajczyk
J Vis Commun Image Represent 2016
pdf

An Objective Deghosting Quality Metric for HDR Images

TasvirEt: A Benchmark Dataset for Automatic Turkish Description Generation from Images
Mesut Erhan Unal, Begum Citamak, Semih Yagcioglu, Aykut Erdem, Erkut Erdem, Nazli Ikizler Cinbis, Ruket Cakici
SIU 2016
pdf (in Turkish) · project page · Turkish captions for Flickr8K dataset

An Objective Deghosting Quality Metric for HDR Images

An Objective Deghosting Quality Metric for HDR Images
Okan Tarhan Tursun, Ahmet Oguz Akyuz, Aykut Erdem, Erkut Erdem
EG 2016
pdf · project page (with code) · experiment page

The State of the Art in HDR Deghosting: A Survey and Evaluation

The State of the Art in HDR Deghosting: A Survey and Evaluation
Okan Tarhan Tursun, Ahmet Oguz Akyuz, Aykut Erdem and Erkut Erdem
EG STAR 2015
pdf · project page

Structure-Preserving Image Smoothing via Region Covariances

Structure-Preserving Image Smoothing via Region Covariances
Levent Karacan, Erkut Erdem and Aykut Erdem
SIGGRAPH Asia 2013
pdf · project page (with code)

Graph Transduction as a Non-Cooperative Game

Graph Transduction as a Non-Cooperative Game
Aykut Erdem and Marcello Pelillo
Neural Comput 2012
pdf · code

all publications

My Erdös number is 3 (via 1. Arun Kumar Jagota, 2. Marcello Pelillo)

Assistant Professor
Computer Vision Laboratory
Department of Computer Engineering
Hacettepe University


aykut@cs.hacettepe.edu.tr
profile
@aykuterdem
aykute

Department of Computer Engineering
Hacettepe University
Beytepe Campus
Ankara, Turkey TR-06800
+90 (312) 780-7545

Office hours: Tuesdays 4-5pm


Curriculum vitae · all publications


Teaching (2016-2017)
Spring BBM202: Algorithms
CMP722: Deep Learning for CV
Fall BBM101: Intro to Programming I
BBM406: Fundamentals of Machine Learning
previous courses


Talks and travel
6 Sep OBAYO
11 Dec NIPS 2016
15 May SIU 2017
21 Aug BYOYO 2017
22 Oct ICCV 2017

I'm looking for motivated MSc/PhD students.
a curated list of resources for grad students

PhD students
Bora Celikkale
Kemal Cizmeciler
Levent Karacan
Yasin Kavak
Aysun Kocak
Semih Yagcioglu

MSc students
Begum Citamak
Mehmet Gunel
Goksu Erdogan

Former Students
Osman Akin (PhD) 2016
Burcak Asal (MSc) 2017
Cagdas Bak (MSc) 2016
Mert Kilickaya (MSc) 2016
Hasan Tugrul Erdogan (MSc) 2015


Our research is made possible through generous support by The Scientific and Technological Research Council of Turkey (TUBITAK) and NVIDIA Corporation.


Some books I've read and enjoyed

Page design courtesy of Michael Bernstein