Loading...
Loading...

Go to the content (press return)

A bio-inspired quaternion local phase CNN layer with contrast invariance and linear sensitivity to rotation angles

Author
Moya-Sánchez, U.; Xambo, S.; Sánchez-Pérez, A.; Salazar-Colores, S.; Martínez-Ortega, J.; Cortes, U.
Type of activity
Journal article
Journal
Pattern recognition letters
Date of publication
2020-03
Volume
131
First page
56
Last page
62
DOI
10.1016/j.patrec.2019.12.001
URL
https://www.sciencedirect.com/science/article/abs/pii/S0167865519303642 Open in new window
Abstract
Deep learning models have been particularly successful with image recognition using Convolutional Neural Networks (CNN). However, the learning of a contrast invariance and rotation equivariance response may fail even with very deep CNNs or by large data augmentations in training. We were inspired by the V1 visual features of the mammalian visual system to emulate as much as possible the early visual system and add more invariant capacities to the CNN. We present a new quaternion local phase conv...
Group of research
IDEAI-UPC - Intelligent Data Science and Artificial Intelligence Research Center
KEMLG - Knowledge Engineering and Machine Learning Group

Participants