DEEP learning-based VIsion and Signal Processing Group

“The only thing worse than being blind is having sight but no vision.”

Helen Keller

Welcome to our Deep Learning-based Vision and Signal Processing Group. Positioned at the forefront of research in deep learning for vision, infrared vision, and various other vision/signal-related domains, our focus is on leveraging extensive experience to delve into the nuances of visual perception. Our commitment extends beyond the realm of research, as we strive to distill complex vision-related concepts into accessible insights. We dedicate ourselves to advancing our understanding of vision through the continued application of cutting-edge techniques in deep learning and related fields.

Principle Investigators

Dr. Erdem Akagündüz is an associate professor with the Graduate School of Informatics (METU) and the principal investigator of the DeepVIS group. His research interests cover computer vision and applied deep learning.

Projects

  • “Deep Learning-based and Real-Time Visual Tracking System Using Infrared and Visible Band Fusion”
    METU Scientific Research Project (BAP) (AGEP-704-11000) October 2022 – January 2024
    Supervisor: Dr. Akagündüz,
    Research Students: Abbas Türkoğlu (MSc Thesis: Deep Learning-based Object Tracking System by Using Visual and Thermal Infrared Fusion)
  • “GPU Driver Infrastructure Construction for Deep Learning Research”
    Çankaya Üni. BAP Infrastructure Project (MF.20.008) September 2020 – August 2021
    Supervisor: Dr. Akagündüz,

Publications

  • (2024) Aytekin Erdoğan, Erdem Akagündüz, “FuseFormer: A Transformer for Visual and Thermal Image Fusion” (under review)
  • (2023) Abbas Türkoğlu, Erdem Akagündüz, “EANet: Enhanced Attribute-Based RGBT Tracker Network”, 16th International Conference on Machine Vision.
  • (2023) Fatih Çağan Akyon, Erdem Akagündüz, Sinan Onur Altınuç, Alptekin Temizel “Sequence Models for Drone vs Bird Classification”, 16th International Conference on Machine Vision.
  • (2023) Abbas Türkoğlu “Deep Learning-based Object Tracking System by Using Visual and Thermal Infrared Fusion” Master Thesis submitted to the Graduate School of Informatics, METU. Supervisor: Dr. Akagündüz.
  • (2023) Aytekin Erdoğan “A Transformer-based Approach for Fusing Infrared and Visible Band Images” Master Thesis submitted to the Graduate School of Informatics, METU. Supervisor: Dr. Akagündüz.
  • (2023) Kevser Irem Danaci, Erdem Akagündüz “A Survey on Infrared Image and Video Sets” J. of Multimedia Tools and Applications.
  • (2023) Kevser İrem Danacı, “Derin Öğrenme Tabanlı Kızılötesi Hedef Tespiti” Master Thesis Submitted to Sivas Science and Technology University, Supervisor: Dr. Akagündüz
  • (2022) Irem Ulku, Erdem Akagündüz, Pedram Ghamisi “Deep Semantic Segmentation of Trees Using Multi-Spectral Images” in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing, vol. 15, pp. 7589-7604, 2022, doi: 10.1109/JSTARS.2022.3203145.
  • (2022) Gülin Tüfekci, Alper Kayabaşı, Erdem Akagündüz, İlkay Ulusoy “Detecting Driver Drowsiness as an Anomaly Using LSTM Autoencoders”, ECCV 2022 Workshops. Lecture Notes in Computer Science, vol 13806. 
  • (2022) Irem Ulku, Erdem Akagündüz, “Semantic Segmentation of Crop Areas in Remote Sensing Imagery using Spectral Indices and Multiple Channels”, 15th International Conference on Machine Vision
  • (2022) “Efficient Implementation of Convolutional Neural Networks on Embedded Devices , Master Thesis submitted to Dept. of Electrical and Electronics Engineering, Çankaya Uni. Supervisor: Dr. Akagündüz
    submitted to the Department of Electrical & Electronics Engineering,
    Çankaya University, Ankara, Turkey
  • (2022) Engin Uzun, Ahmet Anıl Dursun, Erdem Akagündüz, “Augmentation of Atmospheric Turbulence Effects on Thermal Adapted Object Detection Models” 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), New Orleans, LA, USA, 2022, pp. 240-247, doi: 10.1109/CVPRW56347.2022.00038.
  • (2022) Irem Ulku & Erdem Akagündüz “A Survey on Deep Learning-based Architectures for Semantic Segmentation on 2D Images” Applied Artificial Intelligence, 36:1, 
    DOI: 10.1080/08839514.2022.2032924
  • (2020) H. Seçkin Demir, Erdem Akagündüz “Filter design for small target detection on infrared imagery using normalized-cross-correlation layer,” Turkish Journal of Electrical Engineering and Computer Sciences: Vol. 28: No. 1
  • (2019) Erdem Akagunduz, A. G. Bors and K. K. Evans, “Defining Image Memorability Using the Visual Memory Schema,” in IEEE Transactions on Pattern Analysis and Machine Intelligence, vol. 42, no. 9, pp. 2165-2178, 1 Sept. 2020, doi: 10.1109/TPAMI.2019.2914392.
  • (2019) Irem Ulku, Panagiotis Barmpoutis, Tania Stathaki, Erdem Akagündüz, “Comparison of Single Channel Indices for U-Net-based Segmentation of Vegetation in Satellite Images”, 12th International Conference on Machine Vision
  • (2016) Erdem Akagündüz, M. Aslan, A. Şengür, H. Wang and M. C. İnce, “Silhouette Orientation Volumes for Efficient Fall Detection in Depth Videos,” in IEEE Journal of Biomedical and Health Informatics, vol. 21, no. 3, pp. 756-763, May 2017, doi: 10.1109/JBHI.2016.2570300
  • (2015) Erhan Gündoğdu, Hüseyin Özkan, H. Seçkin Demir, Hamza Ergezer, Erdem Akagündüz, S. Kubilay Pakin, “Comparison of infrared and visible imagery for object tracking: Toward trackers with superior IR performance,” 2015 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW), Boston, MA, USA, 2015, pp. 1-9, doi: 10.1109/CVPRW.2015.7301290