Jonas Unger
Professor, Avdelningschef
Avdelningschef för Medie- och informationsteknik
Forskningsledare för Visuell och interaktiv AI
Publikationer
2026
Evaluating representation conditioned diffusion models: A comparative study of image-based representation encoders
Journal of Visual Communication and Image Representation, Vol. 121, Artikel 104950
(Artikel i tidskrift)
https://dx.doi.org/10.1016/j.jvcir.2026.104950
Smaller and Faster 3DGS via Post-Training Dictionary Learning
EG 2026 - POSTERS
(Konferensbidrag)
https://dx.doi.org/10.2312/egp.20261016
Multi-Agent Reinforcement Learning for Conflict Resolution in Air Traffic Control With Delayed Actions
IEEE Open Journal of Intelligent Transportation Systems, Vol. 7, s. 2037-2044
(Artikel i tidskrift)
https://dx.doi.org/10.1109/ojits.2026.3718458
Representation-Conditioned Diffusion Models for Guided Training Data Generation
Synthetic Data for Computer Vision Workshop (SynData4CV)
(Konferensbidrag)
Enhancing out-of-distribution detection with extended logit normalization
Proceedings of IEEE/CVF Conference on Computer vision and Pattern Recognition (CVPR)
(Konferensbidrag)