Two-phase Dynamic Routing for Micro and Macro-level Equivariance in Multi-Column Capsule Networks. (January 2021)

• Record Type: Journal Article
• Authors: Mandal, Bodhisatwa; Sarkhel, Ritesh; Ghosh, Swarnendu; Das, Nibaran; Nasipuri, Mita
• Is Part Of: Pattern recognition. Volume 109(2021)
• Publication Date: 2021-01
• Subjects: Dynamic routing -- Routing by agreement -- Multi-level -- Multi-column neural network -- Capsule networks; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2020.107595
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: We propose two-phase dynamic routing for micro and macro level aggregation of multiple region-specific capsule networks. We demonstrate the superiority of our model over standard capsule networks and state-of-the-art multicolumn architectures. We evaluate our models on 7 publicaly available datasets. Abstract: The capability of multi column convolutional networks in identifying local invariant features helps improve its performance on image classification tasks to a large extent. Suppression of non maximal activations in a convolutional network, however, can lead to loss of valuable information, as scalar activations typically only, encode the presence (or absence) of a feature in an input image, providing no additional information. Capsule networks, on other hand, learn richer representations by propagating non-maximal activations to higher layers, encoding the agreement between neurons at various layers on the presence (or absence) of a feature into a fixed-length vector. Traditional capsule networks, however encodes agreements for micro and macro-level features of an input image with same precedence. Such an uniform agreement protocol can hinder the repsentation capability of a network, especially for datasets that contain objects with independently deformable components. To address this, we propose a novel two-phase dynamic routing protocol that computes agreements between neurons at various layers for micro and macro-level features, following a hierarchical learning paradigm. Experiments on seven publicly available datasets show that a multi-column capsule network that encodes an input image following our routing protocol performs competitively or better than contemporary multi-column convolutional architectures andtraditional capsule networks on a classification task.Implementations of the networks used in this paper have been made available at: github.com/DVLP-CMATERJU/TwoPhaseDynamicRouting.