no code implementations • 8 Dec 2022 • Ibrahim Batuhan Akkaya, Ugur Halici
Instead of employing thresholding on predictions, we introduce a method to weight the gradients calculated from pseudo-labels based on the reliability of the teacher's predictions.
no code implementations • 26 Feb 2022 • Tolga Aksoy, Ugur Halici
Current state-of-the-art one-stage object detectors are limited by treating each image region separately without considering possible relations of the objects.
no code implementations • 11 Jul 2021 • Gorkem Polat, Yesim Dogrusoz Serinagaoglu, Ugur Halici
In this study, we proposed a framework that analyzes CT lung screenings using convolutional neural networks (CNNs) to reduce false positives.
1 code implementation • 14 Jun 2021 • Ibrahim Batuhan Akkaya, Fazil Altinel, Ugur Halici
To this end, we propose a self-training guided adversarial domain adaptation method to promote generalization capabilities of adversarial domain adaptation methods.
no code implementations • 2 Sep 2020 • Ibrahim Batuhan Akkaya, Ugur Halici
In this work, we propose an unsupervised IIT method that preserves the uniform regions after the translation.
1 code implementation • 10 Aug 2019 • Safa Alver, Ugur Halici
Visual face tracking is one of the most important tasks in video surveillance systems.
2 code implementations • 4 Nov 2018 • Gorkem Polat, Ugur Halici, Yesim Serinagaoglu Dogrusoz
3D CNNs are preferred over 2D CNNs because data are in 3D, and 2D convolutional operations may result in information loss.