PT Unknown
AU Ivet Rafegas
   Maria Vanrell
TI Color representation in CNNs: parallelisms with biological vision
BT ICCV Workshop on Mutual Benefits ofr Cognitive and Computer Vision
PY 2017
AB Convolutional Neural Networks (CNNs) trained for object recognition tasks present representational capabilities approaching to primate visual systems [1]. This provides a computational framework to explore how image featuresare efficiently represented. Here, we dissect a trained CNN[2] to study how color is represented. We use a classical methodology used in physiology that is measuring index of selectivity of individual neurons to specific features. We use ImageNet Dataset [20] images and synthetic versionsof them to quantify color tuning properties of artificial neurons to provide a classification of the network population.We conclude three main levels of color representation showing some parallelisms with biological visual systems: (a) a decomposition in a circular hue space to represent single color regions with a wider hue sampling beyond the firstlayer (V2), (b) the emergence of opponent low-dimensional spaces in early stages to represent color edges (V1); and (c) a strong entanglement between color and shape patterns representing object-parts (e.g. wheel of a car), objectshapes (e.g. faces) or object-surrounds configurations (e.g. blue sky surrounding an object) in deeper layers (V4 or IT).
ER