Capsule networks as recurrent models of grouping and segmentation

PLoS Comput Biol. 2020 Jul 21;16(7):e1008017. doi: 10.1371/journal.pcbi.1008017. eCollection 2020 Jul.

Abstract

Classically, visual processing is described as a cascade of local feedforward computations. Feedforward Convolutional Neural Networks (ffCNNs) have shown how powerful such models can be. However, using visual crowding as a well-controlled challenge, we previously showed that no classic model of vision, including ffCNNs, can explain human global shape processing. Here, we show that Capsule Neural Networks (CapsNets), combining ffCNNs with recurrent grouping and segmentation, solve this challenge. We also show that ffCNNs and standard recurrent CNNs do not, suggesting that the grouping and segmentation capabilities of CapsNets are crucial. Furthermore, we provide psychophysical evidence that grouping and segmentation are implemented recurrently in humans, and show that CapsNets reproduce these results well. We discuss why recurrence seems needed to implement grouping and segmentation efficiently. Together, we provide mutually reinforcing psychophysical and computational evidence that a recurrent grouping and segmentation process is essential to understand the visual system and create better models that harness global shape computations.

Publication types

  • Research Support, Non-U.S. Gov't

MeSH terms

  • Algorithms
  • Computational Biology*
  • Computer Simulation
  • Female
  • Humans
  • Image Processing, Computer-Assisted / methods
  • Male
  • Models, Biological
  • Neural Networks, Computer*
  • Normal Distribution
  • Pattern Recognition, Visual*
  • Reproducibility of Results
  • Vision, Ocular*

Grant support

AD was supported by the Swiss National Science Foundation grant n.176153 “Basics of visual processing: from elements to figures”. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.