B. Franceschiello, A. Mashtakov, G. Citti, A. Sarti. Modelling of the Poggendorff Illusion via Sub-Riemannian Geodesics in the Roto-Translation Group

Authors: B. Franceschiello, A. Mashtakov, G. Citti, A. Sarti
Title: Modelling of the Poggendorff Illusion via Sub-Riemannian Geodesics in the Roto-Translation Group
Abstract:

We present a neuro-mathematical model for the well-known Poggendorff illusion, where an illusory contour appears as a geodesic in some given metric, induced in the primary visual cortex V1 by a visual stimulus. Our model extends the cortical based model by Citti and Sarti of perceptual completion in the roto-translation space SE(2), where the functional architecture and neural connectivity of V1 of mammalians is modelled as principal fiber bundle of SE(2) equipped with a sub-Riemannian (SR) metric. We extend the model by taking into account a presence of a visual stimulus (data adaptivity), which is done by including an appropriate external cost modulating the SR-metric. Perceptual curves appear as geodesics, that we compute via SR-Fast Marching.

Citation: Franceschiello B., Mashtakov A., Citti G., Sarti A. (2017) Modelling of the Poggendorff Illusion via Sub-Riemannian Geodesics in the Roto-Translation Group. In: Battiato S., Farinella G., Leo M., Gallo G. (eds) New Trends in Image Analysis and Processing – ICIAP 2017. ICIAP 2017. Lecture Notes in Computer Science, vol 10590. Springer, Cham
Conference: ICIAP 2017: New Trends in Image Analysis and Processing
Place:

Catania, Italy

Year: 2017
Pages: 37-47
File: Download