Nedko Savov
Computer Vision / Machine Learning Specialist

I am a PhD student at INSAIT, Sofia University, supervised by Luc Van Gool.
Broadly speaking, I am interested in teaching computers to understand and generate visual worlds — from videos to interactive environments.
My current research involves vision-based generative modeling with an emphasis on control and interaction. I am interested in world models, controllable video synthesis/editing and data-driven simulators generalizing across many environments.
In the past, I have extensively worked on face analysis and synthesis topics - including deepfake generation and detection, age and expression estimation, face relighting, etc. I have also provided supervision for UvA Master AI students for their computer vision Master’s thesis projects.
Previously, I finished the Artificial Intelligence Master’s at the University of Amsterdam with honors, and a Computer Science Bachelor’s degree at Sofia University, Bulgaria at the top of the class.
selected publications
- Exploration-Driven Generative Interactive EnvironmentsCVPR, 2025
- Leveraging Driver Field-of-View for Multimodal Ego-Trajectory PredictionIn The Thirteenth International Conference on Learning Representations, 2025
- Learning Generative Interactive Environments By Trained Agent ExplorationIn NeurIPS 2024 Workshop on Data-driven and Differentiable Simulations, Surrogates, and Solvers, 2024
- Spatio-temporal Features for Generalized Detection of Deepfake VideosarXiv preprint arXiv:2010.11844, 2020
- Pose and expression robust age estimation via 3D face reconstruction from a single imageIn Proceedings of the IEEE/CVF International Conference on Computer Vision Workshops, 2019