3 papers
cs.CV2026
Faster Training, Fewer Labels: Self-Supervised Pretraining for Fine-Grained BEV Segmentation
Daniel Busch, Christian Bohn, Thomas Kurbiel +3
Dense Bird's Eye View (BEV) semantic maps are central to autonomous driving, yet current multi-camera methods depend on costly, inconsistently annotated BEV ground truth. We addres…
cond-mat.mtrl-sci2026
Precise, efficient and flexible modeling of crystallizing elastomers based on physics-augmented neural networks
Konrad Friedrichs, Franz DammaÃ, Karl A. Kalina +1
We propose a precise and efficient physics-augmented neural network (PANN) to model strain-induced crystallization in rubbery polymers. We demonstrate that the model can be flexibl…
cs.CV2025
Efficient Inter-Task Attention for Multitask Transformer Models
Christian Bohn, Thomas Kurbiel, Klaus Friedrichs +2
In both Computer Vision and the wider Deep Learning field, the Transformer architecture is well-established as state-of-the-art for many applications. For Multitask Learning, howev…