Understanding and Improving CNNs with Complex Structure Tensor: A Biometrics Study

arXiv:2404.15608v1 Announce Type: new
Abstract: Our study provides evidence that CNNs struggle to effectively extract orientation features. We show that the use of Complex Structure Tensor, which contains compact orientation features with certainties, as input to CNNs consistently improves identification accuracy compared to using grayscale inputs alone. Experiments also demonstrated that our inputs, which were provided by mini complex conv-nets, combined with reduced CNN sizes, outperformed full-fledged, prevailing CNN architectures. This suggests that the upfront use of orientation features in …

abstract accuracy arxiv biometrics cnns compact cs.cv evidence extract features identification improving inputs show struggle study tensor type understanding