Êtes-vous un étudiant de l'EPFL à la recherche d'un projet de semestre?

Travaillez avec nous sur des projets en science des données et en visualisation, et déployez votre projet sous forme d'application sur GraphSearch.

Découvrez-en plus sur les applications Graph.

Concept

Réseau neuronal à capsules

Un réseau neuronal à capsules, abrégé CapsNet, est un type de réseau de neurones artificiels du domaine de l'apprentissage automatique dont l'architecture est basée sur celle d'un réseau neuronal convolutif (CNN) mais qui permet une meilleure modélisation relationnelle des motifs extraits par les opérations de convolution. Cette nouvelle approche est une tentative de mimétiser encore plus le fonctionnement des neurones biologiques mais aussi de répondre aux déficiences naturelles du CNN, à savoir que les motifs reconnus par les opérations de convolution sont indépendants les uns des autres et qu'il est ainsi facile de tromper un CNN. Comme pour le CNN, l'objectif principal de cette architecture est l'étude, la compréhension et la classification d'images. Les différentes opérations du réseau vont extraire des caractéristiques spatiales de l'image étudiée, comme des formes ou des motifs mais aussi les mettre en relation les uns par rapport aux autres afin de proposer c

À propos de ce résultat

Cette page est générée automatiquement et peut contenir des informations qui ne sont pas correctes, complètes, à jour ou pertinentes par rapport à votre recherche. Il en va de même pour toutes les autres pages de ce site. Veillez à vérifier les informations auprès des sources officielles de l'EPFL.

Publications associées

Chargement

Personnes associées

Chargement

Unités associées

Chargement

Concepts associés

Chargement

Cours associés

Chargement

Séances de cours associées

Chargement

Résumé

Source officielle

À propos de ce résultat

Publications associées (8)

Publications associées

Chargement

Personnes associées (2)

Personnes associées

Chargement

Unités associées (1)

Unités associées

Chargement

Concepts associés

Chargement

Cours associés (1)

Since 2010 approaches in deep learning have revolutionized fields as diverse as computer vision, machine learning, or artificial intelligence. This course gives a systematic introduction into influential models of deep artificial neural networks, with a focus on Reinforcement Learning.

Cours associés

Chargement

Séances de cours associées (1)

Séances de cours associées

Chargement

Neural Parts: Learning Expressive 3D Shape Abstractions with Invertible Neural Networks

Andreas Geiger, Angelos Katharopoulos, Despoina Paschalidou

Impressive progress in 3D shape extraction led to representations that can capture object geometries with high fidelity. In parallel, primitive-based methods seek to represent objects as semantically consistent part arrangements. However, due to the simplicity of existing primitive representations, these methods fail to accurately reconstruct 3D shapes using a small number of primitives/parts. We address the trade-off between reconstruction quality and number of parts with Neural Parts, a novel 3D primitive representation that defines primitives using an Invertible Neural Network (INN) which implements homeomorphic mappings between a sphere and the target object. The INN allows us to compute the inverse mapping of the homeomorphism, which in turn, enables the efficient computation of both the implicit surface function of a primitive and its mesh, without any additional post-processing. Our model learns to parse 3D objects into semantically consistent part arrangements without any part-level supervision. Evaluations on ShapeNet, D-FAUST and FreiHAND demonstrate that our primitives can capture complex geometries and thus simultaneously achieve geometrically accurate as well as interpretable reconstructions using an order of magnitude fewer primitives than state-of-the-art shape abstraction methods.

IEEE COMPUTER SOC2021

Chargement

How crowding challenges (feedforward) convolutional neural networks

Alban Bornet, Adrien Christophe Doerig, Gregory Francis, Michael Herzog, Ben Henrik Lönnqvist, Lynn Schmittwilken

Are (feedforward) convolutional neural networks (CNNs) good models for the human visual system? Here, we used visual crowding as a well-controlled psychophysical test to probe CNNs. Visual crowding is a ubiquitous breakdown of object recognition in the human visual system, whereby targets become jumbled and unrecognisable in the presence of flanking objects. Humans exhibit several well-documented effects of crowding, such as invariance to size, where the size of the target and flanker letters may be changed without impacting the strength of crowding. We show that feedforward CNNs are unable to reproduce invariance to size, confusion between target and flanker identities, and importantly uncrowding, where paradoxically increasing the number of flankers improves performance. We investigate this phenomenon using a recurrent, neurally inspired model called LAMINART, which we find can reproduce uncrowding as observed in humans. Furthermore, we show that capsule networks, a recurrent family of CNNs with grouping and segmentation mechanisms, outperform any other models of uncrowding to date, demonstrating the importance of grouping and segmentation in mechanisms in visual information processing in general.

2021

Chargement

Capsule networks explain complex spatial processing

Adrien Christophe Doerig, Michael Herzog, Lynn Schmittwilken

Classically, visual processing is described as a cascade of local feedforward computations. This view has received striking support from the success of convolutional neural networks (CNNs). However, CNNs only roughly mimic human vision. For example, CNNs do not take the global spatial configuration of visual elements into account and thus fail at simple tasks such as explaining crowding and uncrowding. In crowding, the perception of a target deteriorates in the presence of neighboring elements. Classically, adding flanking elements is thought to always decreases performance. However, adding flankers even far away from the target can improve performance, depending on the global configuration (uncrowding). We showed previously that no classic model of crowding, including CNNs, can explain uncrowding. Here, we show that capsule networks, a type of deep network combining CNNs and object segmentation, explain both crowding and uncrowding. We trained capsule networks to recognize targets and groups of shapes. There were no crowding/uncrowding stimuli in the training set. When we subsequently tested the network on crowding/uncrowding stimuli, both crowding and uncrowding occurred. We show theoretically how crowding and uncrowding naturally emerge from neural dynamics in capsule networks. These powerful recurrent models offer a new framework to understand previously unexplained experimental results.

SAGE PUBLICATIONS LTD2019

Chargement