Research

Video Anomaly Detection Detecting unexpected events in real-time.	Adversarial Machine Learning Analyzing vulnerabilities of ML algorithms and developing defense mechanisms.	Time Series Forecasting Predicting future values of multivariate time series and detecting anomalies in them.
Hardware-based Authentication RF fingerprinting and PUF signature design for authentication.	Remote Sensing Landscape characterization for environmental monitoring.	Medical AI AI for medical applications.

Video Anomaly Detection

We investigate detecting anomalies in video streams (e.g., for surveillance systems, autonomous vehicles, social media content moderation) by designing state-of-the-art deep neural networks and more recently with LLMs, VLMs, and agentic AI systems.

Sponsor:

NSF

Sample Publications:

Karim, H. and Yilmaz, 2026. SemVAD: Fusing Semantic and Vision Features for Weakly Supervised Video Anomaly Detection. Transactions on Machine Learning Research (TMLR) [pdf]
Mumcu, F., Jones, M., Yilmaz, Y. and Cherian, A., 2025. ComplexVAD: Detecting Interaction Anomalies in Video. Winter Conference on Applications of Computer Vision Workshops (WACVW).[pdf]
Karim, H. and Yilmaz, 2024. Real-Time Weakly Supervised Video Anomaly Detection. IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) [pdf]
Doshi, K. and Yilmaz, Y., 2023. Towards interpretable video anomaly detection. In Proceedings of the IEEE/CVF winter conference on applications of computer vision (WACV).[pdf]
Doshi, K. and Yilmaz, Y., 2021. Online anomaly detection in surveillance videos with asymptotic bound on false alarm rate. Pattern Recognition[pdf]
Doshi, K. and Yilmaz, Y., 2020. Continual learning for anomaly detection in surveillance videos. IEEE/CVF conference on computer vision and pattern recognition workshops.[pdf]

Adversarial Machine Learning

Attackers can deceive AI systems by designing adversarial data samples or cyberattacks. We investigate potential vulnerabilities of various AI systems and propose solutions.

Sponsor:

NSF

Sample Publications:

Mumcu, F. and Yilmaz, Y., 2025. Universal and Efficient Detection of Adversarial Data through Nonuniform Impact on Network Layers. Transactions on Machine Learning Research (TMLR) [pdf]
Karim, H. and Yilmaz, Y., 2025. Invisibility Cloak: Hiding Anomalies in Videos via Adversarial Machine Learning Attacks. Winter Conference on Applications of Computer Vision (WACV).[pdf]
Mumcu, F. and Yilmaz, Y., 2024. Sequential architecture-agnostic black-box attack design and analysis. Pattern Recognition.[pdf]
Mumcu, F. and Yilmaz, Y., 2024. Multimodal attack detection for action recognition models. IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).[pdf]
Mumcu, F. and Yilmaz, Y., 2024. Fast and lightweight vision-language model for adversarial traffic sign detection. Electronics.[pdf]
Mumcu, F., Doshi, K. and Yilmaz, Y., 2022. Adversarial machine learning attacks against video anomaly detection systems. In Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).[pdf]

Time Series Forecasting

Time series forecasting and anomaly detection problems appear in a wide range of applications, such as environmental monitoring, ship route prediction, energy optimization, etc.

Sponsor:

NIFA

Sample Publications:

Murad, M.M.N. and Yilmaz, Y., 2026. Cluster-Aware Causal Mixer for Online Anomaly Detection in Multivariate Time Series. International Conference on Machine Learning (ICML) [pdf]
Murad, M.M.N., Aktukmak, M. and Yilmaz, Y., 2025. Wpmixer: Efficient multi-resolution mixing for long-term time series forecasting. AAAI Conference on Artificial Intelligence [pdf]
Murad, M.M.N., Yirenya-Tawiah, D.K., Weller, T. and Yilmaz, Y., 2025. Multi-Resolution Mixer Network for Localization of Multiple Sensors from Cumulative Power Measurements. IEEE Wireless Communications and Networking Conference (WCNC). [pdf]
Doshi, K., Abudalou, S. and Yilmaz, Y., 2022. Reward once, penalize once: Rectifying time series anomaly detection. In 2022 International Joint Conference on Neural Networks (IJCNN). [pdf]

Hardware-based Authentication

Hardware signatures can significantly complement software-based authentication. We introduce enhanced RF fingerprints and Physically Unclonable Functions (PUFs) through additively manufactured antenna arrays and digital circuits, respectively.

Sponsor:

Army

Sample Publications:

Ranstrom, T., Jebreil, O., Razak, F.A., Yilmaz, Y. and Mumcu, G., 2026. Analytical and Experimental Validation of Wireless Authentication through Enhanced RF Fingerprints of Chaotic Antenna Arrays. IEEE Transactions on Information Forensics and Security. [pdf]
Pendino, A., Nguyen, N., Nouma, S., Wang, J., Yavuz, A., Yilmaz, Y. and Mumcu, G., 2025. Additively Manufactured RF Electronics With Structurally Integrated Physically Unclonable Functions for Wireless System Security. IEEE Access. [pdf]
McMillen, J., Razak, F.A., Mumcu, G. and Yilmaz, Y., 2025. Hardware and Deep Learning-Based Authentication Through Enhanced RF Fingerprints of 3D-Printed Chaotic Antenna Arrays. IEEE Access. [pdf]
McMillen, J., Mumcu, G. and Yilmaz, Y., 2023. Deep learning-based rf fingerprint authentication with chaotic antenna arrays. IEEE Wireless and Microwave Technology Conference (WAMICON). [pdf]

Remote Sensing

We use drone-based lidar and SAR data, as well as satellite-based data, for landscape characterization and environmental monitoring.

Sponsor:

Army Corps of Engineers

Sample Publications:

McMillen, J. and Yilmaz, Y., 2025. FuseForm: Multimodal Transformer for Semantic Segmentation. Winter Conference on Applications of Computer Vision Workshops (WACVW). [pdf]
McMillen, J. and Yilmaz, Y., 2025. SegGen: An Unreal Engine 5 Pipeline for Generating Multimodal Semantic Segmentation Datasets. Sensors. [pdf]

Medical AI

We collaborate with medical doctors and physical theraphists on medical image segmentation, multimodal data fusion, and federated learning.

Sponsor:

Moffitt Cancer Center

Sample Publications:

Ahmed, S., Parker, N., Park, M., Jeong, D., Peres, L.C., Davis, E.W., Permuth, J.B., Siegel, E.M., Schabath, M.B., Yilmaz, Y. and Rasool, G., 2026. Reliable radiologic skeletal muscle area assessment—a biomarker for cancer cachexia diagnosis. Cells. [pdf]
Tripathi, A., Waqas, A., Schabath, M.B., Yilmaz, Y. and Rasool, G., 2025. HONeYBEE: enabling scalable multimodal AI in oncology through foundation model-driven embeddings. npj Digital Medicine. [pdf]
Abudalou, S., Choi, J., Gage, K., Pow-Sang, J., Yilmaz, Y. and Balagurunathan, Y., 2025. Challenges in Using Deep Neural Networks Across Multiple Readers in Delineating Prostate Gland Anatomy. Journal of Imaging Informatics in Medicine. [pdf]
Koutsoubis, N., Waqas, A., Yilmaz, Y., Ramachandran, R.P., Schabath, M.B. and Rasool, G., 2025. Privacy-preserving Federated Learning and Uncertainty Quantification in Medical Imaging. Radiology: Artificial Intelligence. [pdf]
Tripathi, A., Waqas, A., Venkatesan, K., Yilmaz, Y. and Rasool, G., 2024. Building flexible, scalable, and machine learning-ready multimodal oncology datasets. Sensors. [pdf]
Baldeon-Calisto, M., Wei, Z., Abudalou, S., Yilmaz, Y., Gage, K., Pow-Sang, J. and Balagurunathan, Y., 2023. A multi-object deep neural network architecture to detect prostate anatomy in T2-weighted MRI: Performance evaluation. Frontiers in Nuclear Medicine. [pdf]