Mehmet Aygün

I am a first year PhD student at The University of Edinburgh under supervision of Oisin Mac Aodha. My current research focuses on self-supervised correspondence learning and 3D reconstruction from single view images in the wild.

Before, I did my master's at Technical University of Munich (TUM), worked closely with Laura Leal-Taixé on problems related to 3-4D scene understanding.

Email  /  CV  /  Google Scholar  /  GitHub /  LinkedIn

  • Feb 2021, Our paper titled 4D Panoptic Lidar Segmentation is accepted to CVPR, 2021.
  • Oct 2020, Our paper about 3D shape correspondence is accepted to 3DV, 2020.
  • Jun 2018, I received DAAD scholarship for studying in Germany.
  • Oct 2017, at ICCV 2017, our paper received Honorable Mention Award from TASK-CV workshop.

I'm interested in computer vision and machine learning. In particular, I am interested in designing self-supervised algorithms that can capture the 3D representation of highly articulated objects like animals in complex enviorements, and utilize the 3D representations for semantic tasks like semantic correspondence.

4D Panoptic Lidar Segmentation
Mehmet Aygün, Aljosa Osep, Mark Weber, Maxim Maximov, Cyrill Stachniss, Jens Behley and Laura Leal-Taixe
CVPR, 2021

A new method and evaluation metric for the task of 4D panoptic segmentation.

Unsupervised Dense Shape Correspondence using Heat Kernels
Mehmet Aygün, Zorah Lähner and Daniel Cremers
International Conference on 3D Vision (3DV), 2020

An emprical study showing that Heat Kernels can replace Geodesic Matrices for unsupervised dense shape correnspondence problem

Multi Modal Convolutional Neural Networks for Brain Tumor Segmentation
Mehmet Aygün,Yusuf Hüseyin Şahin and Gözde Ünal
arXiv Preprint, 2018

Comparision of different fusion methods for multi modal tumor segmentation using 3D Convolutional Neural Networks.

Exploiting Convolution Filter Patterns for Transfer Learning
Mehmet Aygün, Yusuf Aytar and Hazım Ekenel
ICCV 2017 TASK-CV Workshop

A new regularization method for Convolutional Neural Networks which tries to transfer statistical regularities between CNN architectures for transfer learning using Gausian Mixture Models.

Apparent Age Esimation Using Ensemble of Deep Learning Models
Refik Malli, Mehmet Aygün, and Hazım Ekenel
CVPR 2016 ChaLearn Workshop

An ensemble method for predicting apparent age using deep learning models.

This ubiquitous CS researcher website template spawned from here.
Last updated: January 2018