Alperen Kantarci
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Alperen Kantarci has a total of 22 co-authors. Here are the top co-authors by number of shared publications:
- Hazım Kemal Ekenel
Istanbul Technical University
4 shared publications - Fabio Valerio Massoli
Istituto di Scienza e Tecnologie dell’Informazione “Alessandro Faedo”, CNR, Pisa, Italy
2 shared publications - Fabrizio Falchi
ISTI-CNR
2 shared publications - Ferda Ofli
Qatar Computing Research Institute Hamad Bin Khalifa University Doha Qatar
2 shared publications - Gemma Roig
Goethe University Frankfurt
2 shared publications - Giuseppe Amato
ISTI-CNR
2 shared publications
Latest Publications
- Open Access
- 4 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 1351-1355
Mind the Student: Behavioral and Contextual Cues for Automated Engagement Prediction in Online Learning
- 4 October 2026
- Published by Association for Computing Machinery (ACM)
- p. 1351-1355
Open AccessMind the Student: Behavioral and Contextual Cues for Automated Engagement Prediction in Online Learning
- 14 September 2026
- Published by Springer Nature
- p. 330-341
Evaluation of Randomization Through Style Transfer for Enhanced Domain Generalization
- 14 September 2026
- Published by Springer Nature
- p. 330-341
Evaluation of Randomization Through Style Transfer for Enhanced Domain Generalization
- Open Access
- 2 October 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (24) , 28839-28851
Bias-aware face mask detection dataset
In December 2019, a novel coronavirus (COVID-19) spread so quickly around the world that many countries had to set mandatory face mask rules in public areas to reduce the transmission of the virus. To monitor public adherence, researchers aimed to rapidly develop efficient systems that can detect faces with masks automatically. However, the lack of representative and novel datasets posed challenges for training efficient models. Early attempts to collect face mask datasets did not account for potential race, gender, and age biases. Therefore, the resulting models show inherent biases toward specific race groups, such as Asian or Caucasian. In this work, we present a novel face mask detection dataset that contains images posted on Twitter during the pandemic from around the world. Unlike previous datasets, the proposed Bias-Aware Face Mask Detection (BAFMD) dataset contains more images from underrepresented races and age groups to mitigate the problem of the face mask detection task. We perform experiments to investigate potential biases in widely used face mask detection datasets and illustrate that the BAFMD dataset yields models with better performance and generalization ability. The dataset is publicly available at https://github.com/Alpkant/BAFMD.
- 2 October 2024
- Published by Springer Nature in Multimedia Tools and Applications
- Vol. 84 (24) , 28839-28851
Open AccessBias-aware face mask detection dataset
In December 2019, a novel coronavirus (COVID-19) spread so quickly around the world that many countries had to set mandatory face mask rules in public areas to reduce the transmission of the virus. To monitor public adherence, researchers aimed to rapidly develop efficient systems that can detect faces with masks automatically. However, the lack of representative and novel datasets posed challenges for training efficient models. Early attempts to collect face mask datasets did not account for potential race, gender, and age biases. Therefore, the resulting models show inherent biases toward specific race groups, such as Asian or Caucasian. In this work, we present a novel face mask detection dataset that contains images posted on Twitter during the pandemic from around the world. Unlike previous datasets, the proposed Bias-Aware Face Mask Detection (BAFMD) dataset contains more images from underrepresented races and age groups to mitigate the problem of the face mask detection task. We perform experiments to investigate potential biases in widely used face mask detection datasets and illustrate that the BAFMD dataset yields models with better performance and generalization ability. The dataset is publicly available at https://github.com/Alpkant/BAFMD.
- Open Access
Bias-Aware Face Mask Detection Dataset
In December 2019, a novel coronavirus (COVID-19) spread so quickly around the world that many countries had to set mandatory face mask rules in public areas to reduce the transmission of the virus. To monitor public adherence, researchers aimed to rapidly develop efficient systems that can detect faces with masks automatically. However, the lack of representative and novel datasets proved to be the biggest challenge. Early attempts to collect face mask datasets did not account for potential race, gender, and age biases. Therefore, the resulting models show inherent biases toward specific race groups, such as Asian or Caucasian. In this work, we present a novel face mask detection dataset that contains images posted on Twitter during the pandemic from around the world. Unlike previous datasets, the proposed Bias-Aware Face Mask Detection (BAFMD) dataset contains more images from underrepresented race and age groups to mitigate the problem for the face mask detection task. We perform experiments to investigate potential biases in widely used face mask detection datasets and illustrate that the BAFMD dataset yields models with better performance and generalization ability. The dataset is publicly available at https://github.com/Alpkant/BAFMD.
Open AccessBias-Aware Face Mask Detection Dataset
In December 2019, a novel coronavirus (COVID-19) spread so quickly around the world that many countries had to set mandatory face mask rules in public areas to reduce the transmission of the virus. To monitor public adherence, researchers aimed to rapidly develop efficient systems that can detect faces with masks automatically. However, the lack of representative and novel datasets proved to be the biggest challenge. Early attempts to collect face mask datasets did not account for potential race, gender, and age biases. Therefore, the resulting models show inherent biases toward specific race groups, such as Asian or Caucasian. In this work, we present a novel face mask detection dataset that contains images posted on Twitter during the pandemic from around the world. Unlike previous datasets, the proposed Bias-Aware Face Mask Detection (BAFMD) dataset contains more images from underrepresented race and age groups to mitigate the problem for the face mask detection task. We perform experiments to investigate potential biases in widely used face mask detection datasets and illustrate that the BAFMD dataset yields models with better performance and generalization ability. The dataset is publicly available at https://github.com/Alpkant/BAFMD.
- Open Access
- 18 August 2022
- Published by Institution of Engineering and Technology (IET) in IET Biometrics
- Vol. 11 (5) , 396-406
Deep patch‐wise supervision for presentation attack detection
Face recognition systems have been widely deployed in various applications, such as online banking and mobile payment. However, these systems are vulnerable to face presentation attacks, which are created by people who obtain biometric data covertly from a person or through hacked systems. In order to detect these attacks, convolutional neural networks (CNN)‐based systems have gained significant popularity recently. CNN‐based systems perform very well on intra‐data set experiments, yet they fail to generalise to the data sets that they have not been trained on. This indicates that they tend to memorise data set‐specific spoof traces. To mitigate this problem, the authors propose a Deep Patch‐wise Supervision Presentation Attack Detection (DPS‐PAD) model approach that combines pixel‐wise binary supervision with patch‐based CNN. The authors’ experiments show that the proposed patch‐based method forces the model not to memorise the background information or data set‐specific traces. The authors extensively tested the proposed method on widely used PAD data sets—Replay‐Mobile and OULU‐NPU—and on a real‐world data set that has been collected for real‐world PAD use cases. The proposed approach is found to be superior on challenging experimental setups. Namely, it achieves higher performance on OULU‐NPU protocol 3, 4 and on inter‐data set real‐world experiments.
- 18 August 2022
- Published by Institution of Engineering and Technology (IET) in IET Biometrics
- Vol. 11 (5) , 396-406
Open AccessDeep patch‐wise supervision for presentation attack detection
Face recognition systems have been widely deployed in various applications, such as online banking and mobile payment. However, these systems are vulnerable to face presentation attacks, which are created by people who obtain biometric data covertly from a person or through hacked systems. In order to detect these attacks, convolutional neural networks (CNN)‐based systems have gained significant popularity recently. CNN‐based systems perform very well on intra‐data set experiments, yet they fail to generalise to the data sets that they have not been trained on. This indicates that they tend to memorise data set‐specific spoof traces. To mitigate this problem, the authors propose a Deep Patch‐wise Supervision Presentation Attack Detection (DPS‐PAD) model approach that combines pixel‐wise binary supervision with patch‐based CNN. The authors’ experiments show that the proposed patch‐based method forces the model not to memorise the background information or data set‐specific traces. The authors extensively tested the proposed method on widely used PAD data sets—Replay‐Mobile and OULU‐NPU—and on a real‐world data set that has been collected for real‐world PAD use cases. The proposed approach is found to be superior on challenging experimental setups. Namely, it achieves higher performance on OULU‐NPU protocol 3, 4 and on inter‐data set real‐world experiments.
- 7 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 33 (6) , 2313-2323
MOCCA: Multilayer One-Class Classification for Anomaly Detection
Anomalies are ubiquitous in all scientific fields and can express an unexpected event due to incomplete knowledge about the data distribution or an unknown process that suddenly comes into play and distorts the observations. Usually, due to such events’ rarity, to train deep learning (DL) models on the anomaly detection (AD) task, scientists only rely on “normal” data, i.e., nonanomalous samples. Thus, letting the neural network infer the distribution beneath the input data. In such a context, we propose a novel framework, named multilayer one-class classification (MOCCA), to train and test DL models on the AD task. Specifically, we applied our approach to autoencoders. A key novelty in our work stems from the explicit optimization of the intermediate representations for the task at hand. Indeed, differently from commonly used approaches that consider a neural network as a single computational block, i.e., using the output of the last layer only, MOCCA explicitly leverages the multilayer structure of deep architectures. Each layer’s feature space is optimized for AD during training, while in the test phase, the deep representations extracted from the trained layers are combined to detect anomalies. With MOCCA, we split the training process into two steps. First, the autoencoder is trained on the reconstruction task only. Then, we only retain the encoder tasked with minimizing the $L_{2}$ distance between the output representation and a reference point, the anomaly-free training data centroid, at each considered layer. Subsequently, we combine the deep features extracted at the various trained layers of the encoder model to detect anomalies at inference time. To assess the performance of the models trained with MOCCA, we conduct extensive experiments on publicly available datasets, namely CIFAR10, MVTec AD, and ShanghaiTech. We show that our proposed method reaches comparable or superior performance to state-of-the-art approaches available in the literature. Finally, we provide a model analysis to give insights regarding the benefits of our training procedure.
- 7 December 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Neural Networks and Learning Systems
- Vol. 33 (6) , 2313-2323
MOCCA: Multilayer One-Class Classification for Anomaly Detection
Anomalies are ubiquitous in all scientific fields and can express an unexpected event due to incomplete knowledge about the data distribution or an unknown process that suddenly comes into play and distorts the observations. Usually, due to such events’ rarity, to train deep learning (DL) models on the anomaly detection (AD) task, scientists only rely on “normal” data, i.e., nonanomalous samples. Thus, letting the neural network infer the distribution beneath the input data. In such a context, we propose a novel framework, named multilayer one-class classification (MOCCA), to train and test DL models on the AD task. Specifically, we applied our approach to autoencoders. A key novelty in our work stems from the explicit optimization of the intermediate representations for the task at hand. Indeed, differently from commonly used approaches that consider a neural network as a single computational block, i.e., using the output of the last layer only, MOCCA explicitly leverages the multilayer structure of deep architectures. Each layer’s feature space is optimized for AD during training, while in the test phase, the deep representations extracted from the trained layers are combined to detect anomalies. With MOCCA, we split the training process into two steps. First, the autoencoder is trained on the reconstruction task only. Then, we only retain the encoder tasked with minimizing the $L_{2}$ distance between the output representation and a reference point, the anomaly-free training data centroid, at each considered layer. Subsequently, we combine the deep features extracted at the various trained layers of the encoder model to detect anomalies at inference time. To assess the performance of the models trained with MOCCA, we conduct extensive experiments on publicly available datasets, namely CIFAR10, MVTec AD, and ShanghaiTech. We show that our proposed method reaches comparable or superior performance to state-of-the-art approaches available in the literature. Finally, we provide a model analysis to give insights regarding the benefits of our training procedure.
- Open Access
MOCCA: Multi-Layer One-Class ClassificAtion for Anomaly Detection
Anomalies are ubiquitous in all scientific fields and can express an unexpected event due to incomplete knowledge about the data distribution or an unknown process that suddenly comes into play and distorts observations. Due to such events' rarity, to train deep learning models on the Anomaly Detection (AD) task, scientists only rely on "normal" data, i.e., non-anomalous samples. Thus, letting the neural network infer the distribution beneath the input data. In such a context, we propose a novel framework, named Multi-layer One-Class ClassificAtion (MOCCA),to train and test deep learning models on the AD task. Specifically, we applied it to autoencoders. A key novelty in our work stems from the explicit optimization of intermediate representations for the AD task. Indeed, differently from commonly used approaches that consider a neural network as a single computational block, i.e., using the output of the last layer only, MOCCA explicitly leverages the multi-layer structure of deep architectures. Each layer's feature space is optimized for AD during training, while in the test phase, the deep representations extracted from the trained layers are combined to detect anomalies. With MOCCA, we split the training process into two steps. First, the autoencoder is trained on the reconstruction task only. Then, we only retain the encoder tasked with minimizing the L_2 distance between the output representation and a reference point, the anomaly-free training data centroid, at each considered layer. Subsequently, we combine the deep features extracted at the various trained layers of the encoder model to detect anomalies at inference time. To assess the performance of the models trained with MOCCA, we conduct extensive experiments on publicly available datasets. We show that our proposed method reaches comparable or superior performance to state-of-the-art approaches available in the literature.
Open AccessMOCCA: Multi-Layer One-Class ClassificAtion for Anomaly Detection
Anomalies are ubiquitous in all scientific fields and can express an unexpected event due to incomplete knowledge about the data distribution or an unknown process that suddenly comes into play and distorts observations. Due to such events' rarity, to train deep learning models on the Anomaly Detection (AD) task, scientists only rely on "normal" data, i.e., non-anomalous samples. Thus, letting the neural network infer the distribution beneath the input data. In such a context, we propose a novel framework, named Multi-layer One-Class ClassificAtion (MOCCA),to train and test deep learning models on the AD task. Specifically, we applied it to autoencoders. A key novelty in our work stems from the explicit optimization of intermediate representations for the AD task. Indeed, differently from commonly used approaches that consider a neural network as a single computational block, i.e., using the output of the last layer only, MOCCA explicitly leverages the multi-layer structure of deep architectures. Each layer's feature space is optimized for AD during training, while in the test phase, the deep representations extracted from the trained layers are combined to detect anomalies. With MOCCA, we split the training process into two steps. First, the autoencoder is trained on the reconstruction task only. Then, we only retain the encoder tasked with minimizing the L_2 distance between the output representation and a reference point, the anomaly-free training data centroid, at each considered layer. Subsequently, we combine the deep features extracted at the various trained layers of the encoder model to detect anomalies at inference time. To assess the performance of the models trained with MOCCA, we conduct extensive experiments on publicly available datasets. We show that our proposed method reaches comparable or superior performance to state-of-the-art approaches available in the literature.
- 1 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3518-3526
Offline Signature Verification on Real-World Documents
Research on offline signature verification has explored a large variety of methods on multiple signature datasets, which are collected under controlled conditions. However, these datasets may not fully reflect the characteristics of the signatures in some practical use cases. Real-world signatures extracted from the formal documents may contain different types of occlusions, for example, stamps, company seals, ruling lines, and signature boxes. Moreover, they may have very high intra-class variations, where even genuine signatures resemble forgeries. In this paper, we address a real-world writer independent offline signature verification problem, in which, a bank’s customers’ transaction request documents that contain their occluded signatures are compared with their clean reference signatures. Our proposed method consists of two main components, a stamp cleaning method based on CycleGAN and signature representation based on CNNs. We extensively evaluate different verification setups, fine-tuning strategies, and signature representation approaches to have a thorough analysis of the problem. Moreover, we conduct a human evaluation to show the challenging nature of the problem. We run experiments both on our custom dataset, as well as on the publicly available Tobacco-800 dataset. The experimental results validate the difficulty of offline signature verification on real-world documents. However, by employing the stamp cleaning process, we improve the signature verification performance significantly.
- 1 June 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 3518-3526
Offline Signature Verification on Real-World Documents
Research on offline signature verification has explored a large variety of methods on multiple signature datasets, which are collected under controlled conditions. However, these datasets may not fully reflect the characteristics of the signatures in some practical use cases. Real-world signatures extracted from the formal documents may contain different types of occlusions, for example, stamps, company seals, ruling lines, and signature boxes. Moreover, they may have very high intra-class variations, where even genuine signatures resemble forgeries. In this paper, we address a real-world writer independent offline signature verification problem, in which, a bank’s customers’ transaction request documents that contain their occluded signatures are compared with their clean reference signatures. Our proposed method consists of two main components, a stamp cleaning method based on CycleGAN and signature representation based on CNNs. We extensively evaluate different verification setups, fine-tuning strategies, and signature representation approaches to have a thorough analysis of the problem. Moreover, we conduct a human evaluation to show the challenging nature of the problem. We run experiments both on our custom dataset, as well as on the publicly available Tobacco-800 dataset. The experimental results validate the difficulty of offline signature verification on real-world documents. However, by employing the stamp cleaning process, we improve the signature verification performance significantly.
- Open Access
Thermal to Visible Face Recognition Using Deep Autoencoders
Visible face recognition systems achieve nearly perfect recognition accuracies using deep learning. However, in lack of light, these systems perform poorly. A way to deal with this problem is thermal to visible cross-domain face matching. This is a desired technology because of its usefulness in night time surveillance. Nevertheless, due to differences between two domains, it is a very challenging face recognition problem. In this paper, we present a deep autoencoder based system to learn the mapping between visible and thermal face images. Also, we assess the impact of alignment in thermal to visible face recognition. For this purpose, we manually annotate the facial landmarks on the Carl and EURECOM datasets. The proposed approach is extensively tested on three publicly available datasets: Carl, UND-X1, and EURECOM. Experimental results show that the proposed approach improves the state-of-the-art significantly. We observe that alignment increases the performance by around 2%. Annotated facial landmark positions in this study can be downloaded from the following link: github.com/Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders .
Open AccessThermal to Visible Face Recognition Using Deep Autoencoders
Visible face recognition systems achieve nearly perfect recognition accuracies using deep learning. However, in lack of light, these systems perform poorly. A way to deal with this problem is thermal to visible cross-domain face matching. This is a desired technology because of its usefulness in night time surveillance. Nevertheless, due to differences between two domains, it is a very challenging face recognition problem. In this paper, we present a deep autoencoder based system to learn the mapping between visible and thermal face images. Also, we assess the impact of alignment in thermal to visible face recognition. For this purpose, we manually annotate the facial landmarks on the Carl and EURECOM datasets. The proposed approach is extensively tested on three publicly available datasets: Carl, UND-X1, and EURECOM. Experimental results show that the proposed approach improves the state-of-the-art significantly. We observe that alignment increases the performance by around 2%. Annotated facial landmark positions in this study can be downloaded from the following link: github.com/Alpkant/Thermal-to-Visible-Face-Recognition-Using-Deep-Autoencoders .
- Open Access
Blockchain-Based Electronic Voting System for Elections in Turkey
Traditional elections satisfy neither citizens nor political authorities in recent years. They are not fully secure since it is easy to attack votes. It threatens also privacy and transparency of voters. Additionally, it takes too much time to count the votes. This paper proposes a solution using Blockchain to eliminate all the disadvantages of conventional elections. Security and data integrity of votes are absolutely provided theoretically. Voter privacy is another requirement that is ensured in the system. Lastly, the waiting time for results decreased significantly in the proposed Blockchain voting system.
Open AccessBlockchain-Based Electronic Voting System for Elections in Turkey
Traditional elections satisfy neither citizens nor political authorities in recent years. They are not fully secure since it is easy to attack votes. It threatens also privacy and transparency of voters. Additionally, it takes too much time to count the votes. This paper proposes a solution using Blockchain to eliminate all the disadvantages of conventional elections. Security and data integrity of votes are absolutely provided theoretically. Voter privacy is another requirement that is ensured in the system. Lastly, the waiting time for results decreased significantly in the proposed Blockchain voting system.