2 papers
cs.CV2024
Harmformer: Harmonic Networks Meet Transformers for Continuous Roto-Translation Equivariance
Tomáš Karella, Adam Harmanec, Jan Kotera +2
CNNs exhibit inherent equivariance to image translation, leading to efficient parameter and data usage, faster learning, and improved robustness. The concept of translation equivar…
cs.CV2023
H-NeXt: The next step towards roto-translation invariant networks
Tomas Karella, Filip Sroubek, Jan Flusser +2
The widespread popularity of equivariant networks underscores the significance of parameter efficient models and effective use of training data. At a time when robustness to unseen…