Sajad Amouei Sheshkal
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
Sajad Amouei Sheshkal has a total of 15 co-authors. Here are the top co-authors by number of shared publications:
- Michael Alexander Riegler
SimulaMet, Oslo, Norway
3 shared publications - Dag Johansen
UiT The Arctic University of Norway, Tromsø, Norway
2 shared publications - Pegah Salehi
SimulaMet, 0167 Oslo, Norway
2 shared publications - Pål Halvorsen
Oslo Metropolitan University, Oslo, Norway
2 shared publications - Saeed Shafiee Sabet
ForzaSys AS, Oslo, Norway
2 shared publications - Sushant Gautam
Department of Computer Science, Oslo Metropolitan University, Oslo, Norway
2 shared publications
Latest Publications
- Open Access
- 4 March 2025
- Published by MDPI AG in Big Data and Cognitive Computing
- Vol. 9 (3) , 59
Comparative Analysis of Audio Feature Extraction for Real-Time Talking Portrait Synthesis
This paper explores advancements in real-time talking-head generation, focusing on overcoming challenges in Audio Feature Extraction (AFE), which often introduces latency and limits responsiveness in real-time applications. To address these issues, we propose and implement a fully integrated system that replaces conventional AFE models with OpenAI’s Whisper, leveraging its encoder to optimize processing and improve overall system efficiency. Our evaluation of two open-source real-time models across three different datasets shows that Whisper not only accelerates processing but also improves specific aspects of rendering quality, resulting in more realistic and responsive talking-head interactions. Although interviewer training systems are considered a potential application, the primary contribution of this work is the improvement of the technical foundations necessary for creating responsive AI avatars. These advancements enable more immersive interactions and expand the scope of AI-driven applications, including educational tools and simulated training environments.
- 4 March 2025
- Published by MDPI AG in Big Data and Cognitive Computing
- Vol. 9 (3) , 59
Open AccessComparative Analysis of Audio Feature Extraction for Real-Time Talking Portrait Synthesis
This paper explores advancements in real-time talking-head generation, focusing on overcoming challenges in Audio Feature Extraction (AFE), which often introduces latency and limits responsiveness in real-time applications. To address these issues, we propose and implement a fully integrated system that replaces conventional AFE models with OpenAI’s Whisper, leveraging its encoder to optimize processing and improve overall system efficiency. Our evaluation of two open-source real-time models across three different datasets shows that Whisper not only accelerates processing but also improves specific aspects of rendering quality, resulting in more realistic and responsive talking-head interactions. Although interviewer training systems are considered a potential application, the primary contribution of this work is the improvement of the technical foundations necessary for creating responsive AI avatars. These advancements enable more immersive interactions and expand the scope of AI-driven applications, including educational tools and simulated training environments.
- Open Access
- 29 November 2024
- Published by MDPI AG in Diagnostics
- Vol. 14 (23) , 2696
Classifying Dry Eye Disease Patients from Healthy Controls Using Machine Learning and Metabolomics Data
Background: Dry eye disease is a common disorder of the ocular surface, leading patients to seek eye care. Clinical signs and symptoms are currently used to diagnose dry eye disease. Metabolomics, a method for analyzing biological systems, has been found helpful in identifying distinct metabolites in patients and in detecting metabolic profiles that may indicate dry eye disease at early stages. In this study, we explored the use of machine learning and metabolomics data to identify cataract patients who suffer from dry eye disease, a topic that, to our knowledge, has not been previously explored. As there is no one-size-fits-all machine learning model for metabolomics data, choosing the most suitable model can significantly affect the quality of predictions and subsequent metabolomics analyses. Methods: To address this challenge, we conducted a comparative analysis of eight machine learning models on two metabolomics data sets from cataract patients with and without dry eye disease. The models were evaluated and optimized using nested k-fold cross-validation. To assess the performance of these models, we selected a set of suitable evaluation metrics tailored to the data set’s challenges. Results: The logistic regression model overall performed the best, achieving the highest area under the curve score of , balanced accuracy of , Matthew’s correlation coefficient of , an F1-score of , and a specificity of . Additionally, following the logistic regression, the XGBoost and Random Forest models also demonstrated good performance. Conclusions: The results show that the logistic regression model with L2 regularization can outperform more complex models on an imbalanced data set with a small sample size and a high number of features, while also avoiding overfitting and delivering consistent performance across cross-validation folds. Additionally, the results demonstrate that it is possible to identify dry eye in cataract patients from tear film metabolomics data using machine learning models.- 29 November 2024
- Published by MDPI AG in Diagnostics
- Vol. 14 (23) , 2696
Open AccessClassifying Dry Eye Disease Patients from Healthy Controls Using Machine Learning and Metabolomics Data
Background: Dry eye disease is a common disorder of the ocular surface, leading patients to seek eye care. Clinical signs and symptoms are currently used to diagnose dry eye disease. Metabolomics, a method for analyzing biological systems, has been found helpful in identifying distinct metabolites in patients and in detecting metabolic profiles that may indicate dry eye disease at early stages. In this study, we explored the use of machine learning and metabolomics data to identify cataract patients who suffer from dry eye disease, a topic that, to our knowledge, has not been previously explored. As there is no one-size-fits-all machine learning model for metabolomics data, choosing the most suitable model can significantly affect the quality of predictions and subsequent metabolomics analyses. Methods: To address this challenge, we conducted a comparative analysis of eight machine learning models on two metabolomics data sets from cataract patients with and without dry eye disease. The models were evaluated and optimized using nested k-fold cross-validation. To assess the performance of these models, we selected a set of suitable evaluation metrics tailored to the data set’s challenges. Results: The logistic regression model overall performed the best, achieving the highest area under the curve score of , balanced accuracy of , Matthew’s correlation coefficient of , an F1-score of , and a specificity of . Additionally, following the logistic regression, the XGBoost and Random Forest models also demonstrated good performance. Conclusions: The results show that the logistic regression model with L2 regularization can outperform more complex models on an imbalanced data set with a small sample size and a high number of features, while also avoiding overfitting and delivering consistent performance across cross-validation folds. Additionally, the results demonstrate that it is possible to identify dry eye in cataract patients from tear film metabolomics data using machine learning models. - Open Access
Comparative Analysis of Audio Feature Extraction for Real-Time Talking Portrait Synthesis
This paper examines the integration of real-time talking-head generation for interviewer training, focusing on overcoming challenges in Audio Feature Extraction (AFE), which often introduces latency and limits responsiveness in real-time applications. To address these issues, we propose and implement a fully integrated system that replaces conventional AFE models with Open AI's Whisper, leveraging its encoder to optimize processing and improve overall system efficiency. Our evaluation of two open-source real-time models across three different datasets shows that Whisper not only accelerates processing but also improves specific aspects of rendering quality, resulting in more realistic and responsive talking-head interactions. These advancements make the system a more effective tool for immersive, interactive training applications, expanding the potential of AI-driven avatars in interviewer training.
Open AccessComparative Analysis of Audio Feature Extraction for Real-Time Talking Portrait Synthesis
This paper examines the integration of real-time talking-head generation for interviewer training, focusing on overcoming challenges in Audio Feature Extraction (AFE), which often introduces latency and limits responsiveness in real-time applications. To address these issues, we propose and implement a fully integrated system that replaces conventional AFE models with Open AI's Whisper, leveraging its encoder to optimize processing and improve overall system efficiency. Our evaluation of two open-source real-time models across three different datasets shows that Whisper not only accelerates processing but also improves specific aspects of rendering quality, resulting in more realistic and responsive talking-head interactions. These advancements make the system a more effective tool for immersive, interactive training applications, expanding the potential of AI-driven avatars in interviewer training.
- Open Access
- 29 October 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 463-468
An Improved Person Re-identification Method by light-weight convolutional neural network
Person Re-identification is defined as a recognizing process where the person is observed by non-overlapping cameras at different places. In the last decade, the rise in the applications and importance of Person Re-identification for surveillance systems popularized this subject in different areas of computer vision. Person Re-identification is faced with challenges such as low resolution, varying poses, illumination, background clutter, and occlusion, which could affect the result of recognizing process. The present paper aims to improve Person Re-identification using transfer learning and application of verification loss function within the framework of Siamese network. The Siamese network receives image pairs as inputs and extract their features via a pre-trained model. EfficientNet was employed to obtain discriminative features and reduce the demands for data. The advantages of verification loss were used in the network learning. Experiments showed that the proposed model performs better than state-of-the-art methods on the CUHK01 dataset. For example, rank5 accuracies are 95.2% (+5.7) for the CUHK01 datasets. It also achieved an acceptable percentage in Rank 1. Because of the small size of the pre-trained model parameters, learning speeds up and there will be a need for less hardware and data.
- 29 October 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 463-468
Open AccessAn Improved Person Re-identification Method by light-weight convolutional neural network
Person Re-identification is defined as a recognizing process where the person is observed by non-overlapping cameras at different places. In the last decade, the rise in the applications and importance of Person Re-identification for surveillance systems popularized this subject in different areas of computer vision. Person Re-identification is faced with challenges such as low resolution, varying poses, illumination, background clutter, and occlusion, which could affect the result of recognizing process. The present paper aims to improve Person Re-identification using transfer learning and application of verification loss function within the framework of Siamese network. The Siamese network receives image pairs as inputs and extract their features via a pre-trained model. EfficientNet was employed to obtain discriminative features and reduce the demands for data. The advantages of verification loss were used in the network learning. Experiments showed that the proposed model performs better than state-of-the-art methods on the CUHK01 dataset. For example, rank5 accuracies are 95.2% (+5.7) for the CUHK01 datasets. It also achieved an acceptable percentage in Rank 1. Because of the small size of the pre-trained model parameters, learning speeds up and there will be a need for less hardware and data.