TY  - JOUR
U1  - Zeitschriftenartikel, wissenschaftlich - nicht begutachtet (unreviewed)
A1  - Spravil, Julian
A1  - Houben, Sebastian
A1  - Behnke, Sven
T1  - HyenaPixel: Global Image Context with Convolutions
N2  - In vision tasks, a larger effective receptive field (ERF) is associated with better performance. While attention natively supports global context, convolution requires multiple stacked layers and a hierarchical structure for large context. In this work, we extend Hyena, a convolution-based attention replacement, from causal sequences to the non-causal two-dimensional image space. We scale the Hyena convolution kernels beyond the feature map size up to 191$\times$191 to maximize the ERF while maintaining sub-quadratic complexity in the number of pixels. We integrate our two-dimensional Hyena, HyenaPixel, and bidirectional Hyena into the MetaFormer framework. For image categorization, HyenaPixel and bidirectional Hyena achieve a competitive ImageNet-1k top-1 accuracy of 83.0% and 83.5%, respectively, while outperforming other large-kernel networks. Combining HyenaPixel with attention further increases accuracy to 83.6%. We attribute the success of attention to the lack of spatial bias in later stages and support this finding with bidirectional Hyena.
AX  - 2402.19305
SP  - 13
S1  - 13
PB  - arXiv
ER  -