Visual Features and Their Own Optical Flow

Symmetries, invariances and conservation equations have always been an invaluable guide in Science to model natural phenomena through simple yet effective relations. For instance, in computer vision, translation equivariance is typically a built-in property of neural architectures that are used to s...

Descripción completa

Guardado en:

Detalles Bibliográficos
Autores principales:	Alessandro Betti, Giuseppe Boccignone, Lapo Faggi, Marco Gori, Stefano Melacci
Formato:	article
Lenguaje:	EN
Publicado:	Frontiers Media S.A. 2021
Materias:	affordance convolutional neural networks feature flow motion invariance optical flow transport equation Electronic computers. Computer science QA75.5-76.95
Acceso en línea:	https://doaj.org/article/43f817fbb0154a44b9ec3352a84b2bd1
Etiquetas:	Agregar Etiqueta Sin Etiquetas, Sea el primero en etiquetar este registro!

id	oai:doaj.org-article:43f817fbb0154a44b9ec3352a84b2bd1
record_format	dspace
spelling	oai:doaj.org-article:43f817fbb0154a44b9ec3352a84b2bd12021-12-02T00:35:51ZVisual Features and Their Own Optical Flow2624-821210.3389/frai.2021.768516https://doaj.org/article/43f817fbb0154a44b9ec3352a84b2bd12021-12-01T00:00:00Zhttps://www.frontiersin.org/articles/10.3389/frai.2021.768516/fullhttps://doaj.org/toc/2624-8212Symmetries, invariances and conservation equations have always been an invaluable guide in Science to model natural phenomena through simple yet effective relations. For instance, in computer vision, translation equivariance is typically a built-in property of neural architectures that are used to solve visual tasks; networks with computational layers implementing such a property are known as Convolutional Neural Networks (CNNs). This kind of mathematical symmetry, as well as many others that have been recently studied, are typically generated by some underlying group of transformations (translations in the case of CNNs, rotations, etc.) and are particularly suitable to process highly structured data such as molecules or chemical compounds which are known to possess those specific symmetries. When dealing with video streams, common built-in equivariances are able to handle only a small fraction of the broad spectrum of transformations encoded in the visual stimulus and, therefore, the corresponding neural architectures have to resort to a huge amount of supervision in order to achieve good generalization capabilities. In the paper we formulate a theory on the development of visual features that is based on the idea that movement itself provides trajectories on which to impose consistency. We introduce the principle of Material Point Invariance which states that each visual feature is invariant with respect to the associated optical flow, so that features and corresponding velocities are an indissoluble pair. Then, we discuss the interaction of features and velocities and show that certain motion invariance traits could be regarded as a generalization of the classical concept of affordance. These analyses of feature-velocity interactions and their invariance properties leads to a visual field theory which expresses the dynamical constraints of motion coherence and might lead to discover the joint evolution of the visual features along with the associated optical flows.Alessandro BettiGiuseppe BoccignoneLapo FaggiLapo FaggiMarco GoriMarco GoriStefano MelacciFrontiers Media S.A.articleaffordanceconvolutional neural networksfeature flowmotion invarianceoptical flowtransport equationElectronic computers. Computer scienceQA75.5-76.95ENFrontiers in Artificial Intelligence, Vol 4 (2021)
institution	DOAJ
collection	DOAJ
language	EN
topic	affordance convolutional neural networks feature flow motion invariance optical flow transport equation Electronic computers. Computer science QA75.5-76.95
spellingShingle	affordance convolutional neural networks feature flow motion invariance optical flow transport equation Electronic computers. Computer science QA75.5-76.95 Alessandro Betti Giuseppe Boccignone Lapo Faggi Lapo Faggi Marco Gori Marco Gori Stefano Melacci Visual Features and Their Own Optical Flow
description	Symmetries, invariances and conservation equations have always been an invaluable guide in Science to model natural phenomena through simple yet effective relations. For instance, in computer vision, translation equivariance is typically a built-in property of neural architectures that are used to solve visual tasks; networks with computational layers implementing such a property are known as Convolutional Neural Networks (CNNs). This kind of mathematical symmetry, as well as many others that have been recently studied, are typically generated by some underlying group of transformations (translations in the case of CNNs, rotations, etc.) and are particularly suitable to process highly structured data such as molecules or chemical compounds which are known to possess those specific symmetries. When dealing with video streams, common built-in equivariances are able to handle only a small fraction of the broad spectrum of transformations encoded in the visual stimulus and, therefore, the corresponding neural architectures have to resort to a huge amount of supervision in order to achieve good generalization capabilities. In the paper we formulate a theory on the development of visual features that is based on the idea that movement itself provides trajectories on which to impose consistency. We introduce the principle of Material Point Invariance which states that each visual feature is invariant with respect to the associated optical flow, so that features and corresponding velocities are an indissoluble pair. Then, we discuss the interaction of features and velocities and show that certain motion invariance traits could be regarded as a generalization of the classical concept of affordance. These analyses of feature-velocity interactions and their invariance properties leads to a visual field theory which expresses the dynamical constraints of motion coherence and might lead to discover the joint evolution of the visual features along with the associated optical flows.
format	article
author	Alessandro Betti Giuseppe Boccignone Lapo Faggi Lapo Faggi Marco Gori Marco Gori Stefano Melacci
author_facet	Alessandro Betti Giuseppe Boccignone Lapo Faggi Lapo Faggi Marco Gori Marco Gori Stefano Melacci
author_sort	Alessandro Betti
title	Visual Features and Their Own Optical Flow
title_short	Visual Features and Their Own Optical Flow
title_full	Visual Features and Their Own Optical Flow
title_fullStr	Visual Features and Their Own Optical Flow
title_full_unstemmed	Visual Features and Their Own Optical Flow
title_sort	visual features and their own optical flow
publisher	Frontiers Media S.A.
publishDate	2021
url	https://doaj.org/article/43f817fbb0154a44b9ec3352a84b2bd1
work_keys_str_mv	AT alessandrobetti visualfeaturesandtheirownopticalflow AT giuseppeboccignone visualfeaturesandtheirownopticalflow AT lapofaggi visualfeaturesandtheirownopticalflow AT lapofaggi visualfeaturesandtheirownopticalflow AT marcogori visualfeaturesandtheirownopticalflow AT marcogori visualfeaturesandtheirownopticalflow AT stefanomelacci visualfeaturesandtheirownopticalflow
_version_	1718403611861450752

Visual Features and Their Own Optical Flow

Ejemplares similares