Cross channel weight sharing for image classification

Open Webpage

Nazmul Shahadat, Anthony S. Maida

Published: 01 Jan 2024, Last Modified: 05 Nov 2025Image Vis. Comput. 2024EveryoneRevisionsBibTeXCC BY-SA 4.0
Abstract: Highlights•Hypercomplex CNNs with and without attention introduce cross-channel weight sharing.•Cross-channel weight sharing enables discovery of a new class of input correlations.•Hypercomplex networks can be further optimized without loss of accuracy.•Toward this goal, separable hypercomplex operations are introduced.