Welcome to Shervin R. Arashloo's website!

I am a faculty member in the Department of Computer Engineering at Bilkent University.

My research centres on the intersection of computer vision, machine learning, and data science, addressing mathematical and learning foundations of trustworthy visual AI, with a focus on one-class classification, deep and kernel learning, and biometric security (face PAD, deepfake detection, adversarial defense).

I serve as Associate Editor for Pattern Recognition, IEEE TCSVT, and Neurocomputing, and have sustained international collaborations with leading research groups (CVSSP, University of Surrey, and CSEE, University of Essex, UK).

Interests:

• Develop data/label-efficient machine learning methods targeting robustness to unseen conditions and domain shifts.

• Advance algorithmic robustness by designing methods resilient to spoofing, deepfake manipulations, and adversarial attacks.

• Translate kernel-based and one-class theoretical advances into scalable deep architectures and deployable models for visual recognition tasks.