Corpus ID: 173990830

Sample-efficient image segmentation through recurrence

  title={Sample-efficient image segmentation through recurrence},
  author={Drew Linsley and Junkyung Kim and David Berson},
  journal={arXiv: Computer Vision and Pattern Recognition},
There is a growing consensus in vision science that recurrent neural networks constitute better models of visual cortex than feedforward architectures. Yet, feedforward neural networks continue to dominate most popular computer vision challenges. We bridge this gap with the Gamma-net. Inspired by recurrent feedback loops prevalent in the mammalian visual cortex, Gamma-net introduces gated recurrent dynamics through feedforward, horizontal, and top-down connections into the popular U-Net… Expand
8 Citations
Capsule networks as recurrent models of grouping and segmentation
  • 4
  • PDF
Capsule Networks as Recurrent Models ofGrouping and Segmentation
  • 3
  • PDF
Recurrence is required to capture the representational dynamics of the human visual system
  • 85
  • PDF
Beyond the feedforward sweep: feedback computations in the visual cortex
  • G. Kreiman, T. Serre
  • Computer Science, Medicine
  • Annals of the New York Academy of Sciences
  • 2020
  • 13
  • PDF
Disentangling neural mechanisms for perceptual grouping
  • 17
  • PDF
Crowding Reveals Fundamental Differences in Local vs. Global Processing in Humans and Machines
  • 10
  • PDF
Crowding and the architecture of the visual system


Learning long-range spatial dependencies with horizontal gated-recurrent units
  • 34
  • PDF
Task-Driven Convolutional Recurrent Models of the Visual System
  • 83
  • PDF
Recurrent Convolutional Neural Networks: A Better Model of Biological Object Recognition
  • 99
  • PDF
Recurrent convolutional neural network for object recognition
  • Ming Liang, Xiaolin Hu
  • Computer Science
  • 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)
  • 2015
  • 640
  • PDF
Recurrent computations for visual pattern completion
  • 99
  • PDF
Reinforcement Learning of Linking and Tracing Contours in Recurrent Neural Networks
  • 28
  • PDF
Multi-stage Multi-recursive-input Fully Convolutional Networks for Neuronal Boundary Detection
  • 40
  • PDF