Emre Akbaş

E-mail

eakbas@metu.edu.tr

Department

Department of Computer Engineering

ORCID

0000-0002-3760-6722

Scopus Author ID

15123935000

Web of Science Researcher ID

R-4173-2017

Transformers in Small Object Detection: A Benchmark and Survey of State-of-the-Art Miri Rekavandi, Aref; Rashidi, Shima; Boussaid, Farid; Hoefs, Stephen; Akbaş, Emre; Bennamoun, Mohammed (2025-09-10) Transformers have rapidly gained popularity in computer vision, especially in the field of object detection. Upon examining the outcomes of state-of-the-art object detection methods, we noticed that transformers consistent...
WAIT: Feature warping for animation to illustration video translation using GANs Hicsonmez, Samet; Samet, Nermin; Samet, Fidan; Bakir, Oguz; Akbaş, Emre; DUYGULU ŞAHİN, PINAR (2025-07-07) In this paper, we explore a new domain for video-to-video translation. Motivated by the availability of animation movies that are adopted from illustrated books for children, we aim to stylize these videos with the style o...
Colorectal cancer tumor grade segmentation: A new dataset and baseline results Arslan, Duygu; Sehlaver, Sina; Guder, Erce; Temena, Mehmet Arda; Bahcekapili, Alper; Ozdemir, Umut; Turkay, Duriye Ozer; Guner, Gunes; Guresci, Servet; SÖKMENSÜER, CENK; Akbaş, Emre; Acar, Ahmet (2025-02-28) Routine pathology assessment for the tumor grading is currently performed under the microscope by experienced pathologists which might be prone to interpersonal variability and requiring years of experience. Over the past ...
Bucketed Ranking-Based Losses for Efficient Training of Object Detectors Yavuz, Feyza; Cam, Baris Can; Doğan, Adnan Harun; Oksuz, Kemal; Akbaş, Emre; KALKAN, SİNAN (2025-01-01) Ranking-based loss functions, such as Average Precision Loss and Rank&Sort Loss, outperform widely used score-based losses in object detection. These loss functions better align with the evaluation criteria, have fewer hyp...
DeepKin: Predicting Relatedness From Low-Coverage Genomes and Palaeogenomes With Convolutional Neural Networks Güler, Murat; Yılmaz, Ardan; Katırcıoğlu, Büşra; Kantar, Sarp; Ünver, Tara Ekin; Vural, Kıvılcım Başak; ALTINIŞIK, NEFİZE EZGİ; Akbaş, Emre; Somel, Mehmet (2025-01-01) DeepKin is a novel tool designed to predict relatedness from genomic data using convolutional neural networks (CNNs). Traditional methods for estimating relatedness often struggle when genomic data is limited, as with pala...
AIDCON: An Aerial Image Dataset and Benchmark for Construction Machinery Ersöz, Ahmet Bahaddin; Pekcan, Onur; Akbaş, Emre (2024-09-01) Applying deep learning algorithms in the construction industry holds tremendous potential for enhancing site management, safety, and efficiency. The development of such algorithms necessitates a comprehensive and diverse i...
SegIns: A simple extension to instance discrimination task for better localization learning Baydar, Melih; Akbaş, Emre (2024-04-01) Recent self-supervised learning methods, where instance discrimination task is a fundamental way of pretraining convolutional neural networks (CNN), excel in transfer learning performance. Even though instance discriminati...
Dense depth alignment for human pose and shape estimation Karagoz, Batuhan; Suat, Ozhan; Uguz, Bedirhan; Akbaş, Emre (2024-01-01) Estimating 3D human pose and shape (HPS) from a monocular image has many applications. However, collecting ground-truth data for this problem is costly and constrained to limited lab environments. Researchers have used pri...
A multi-level multi-label text classification dataset of 19th century Ottoman and Russian literary and critical texts Gokceoglu, Gokcen; Cavusoglu, Devrim; Akbaş, Emre; Dolcerocca, Ozen Nergis (2024-01-01) This paper introduces a multi-level, multi-label text classification dataset comprising over 3000 documents. The dataset features literary and critical texts from 19th-century Ottoman Turkish and Russian. It is the first s...
Correlation Loss: Enforcing Correlation between Classification and Localization Kahraman, Fehmi; Oksuz, Kemal; KALKAN, SİNAN; Akbaş, Emre (2023-06-27) Object detectors are conventionally trained by a weighted sum of classification and localization losses. Recent studies (e.g., predicting IoU with an auxiliary head, Generalized Focal Loss, Rank & Sort Loss) have shown tha...

Citation Formats