• Corpus ID: 235422265

On-Off Center-Surround Receptive Fields for Accurate and Robust Image Classification

  title={On-Off Center-Surround Receptive Fields for Accurate and Robust Image Classification},
  author={Zahra Babaiee and Ramin M. Hasani and Mathias Lechner and Daniela Rus and Radu Grosu},
Robustness to variations in lighting conditions is a key objective for any deep vision system. To this end, our paper extends the receptive field of convolutional neural networks with two residual components, ubiquitous in the visual processing system of vertebrates: On-center and off-center pathways, with excitatory center and inhibitory surround; OOCS for short. The on-center pathway is excited by the presence of a light stimulus in its center, but not in its surround, whereas the off-center… 
A precortical module for robust CNNs to light variations
The analogies between the cortical level of the visual system and the structure of popular CNNs suggests the introduction of an additional preliminary convolutional module inspired to precortical neuronal circuits to improve robustness with respect to global light intensity and contrast variations in the input images.
3D-OOCS: Learning Prostate Segmentation with Inductive Bias
This work intro-duce OOCS-enhanced networks, a novel architecture inspired by the innate nature of visual processing in the vertebrates that helps 3D U-Nets to scrutinise and delineate anatomical structures present in 3D images with increased accuracy.
Interactive Analysis of CNN Robustness
Perturber is a web‐based application that allows users to instantaneously explore how CNN activations and predictions evolve when a 3D input scene is interactively perturbed, and replicate users’ insights with other CNN architectures and input images, yielding new insights about the vulnerability of adversarially trained models.
GoTube: Scalable Stochastic Verification of Continuous-Depth Models
This work introduces a new stochastic verification algorithm that formally quantifies the behavioral robustness of any timecontinuous process formulated as a continuous-depth model and calls it GoTube, which is stable and sets the state of the art in terms of its ability to scale to time horizons well beyond what has been previously possible.


A Simple Computational Model of Center-Surround Receptive Fields in the Retina
A simple computational model of the outer retinal layers is disclosed that incorporates both rectified on-center/off-surround and off- center/on-Surround bipolar elements, and an increase in the convergence of receptors to output elements that accounts for resolution differences between central and peripheral vision.
Non-linear Convolution Filters for CNN-Based Learning
This work addresses the issue of developing a convolution method in the context of a computational model of the visual cortex, exploring quadratic forms through the Volterra kernels, and shows that a network which combines linear and non-linear filters in its convolutional layers, can outperform networks that use standard linear filters with the same architecture.
Convolutional Neural Network with Biologically Inspired Retinal Structure
Correspondence of Deep Neural Networks and the Brain for Visual Textures
This work developed a quantitative approach for selecting a subset of the neural unit population from the CNN that best describes the brain neural recordings, and found that the first two layers of the CNN showed qualitative and quantitative correspondence to the cortical data across a number of metrics.
Selective Kernel Networks
Detailed analyses show that the neurons in SKNet can capture target objects with different scales, which verifies the capability of neurons for adaptively adjusting their receptive field sizes according to the input.
Stimulus specific responses from beyond the classical receptive field: neurophysiological mechanisms for local-global comparisons in visual neurons.
The historical development of the evidence of response selectivity for visual stimuli presented beyond the CRF is traced; the anatomical pathways that sub serve these far-reaching surround mechanisms are examined; and the possible relationships between these mechanisms and perception are explored.
Understanding the Effective Receptive Field in Deep Convolutional Neural Networks
The notion of an effective receptive fieldsize is introduced, and it is shown that it both has a Gaussian distribution and only occupies a fraction of the full theoretical receptive field size.
Surround Modulation: A Bio-inspired Connectivity Structure for Convolutional Neural Networks
Inspired by the notion of surround modulation, new excitatory-inhibitory connections between a unit and its surrounding units in the convolutional neural network (CNN) are designed to achieve a more biologically plausible network.
Neuronal responses to static texture patterns in area V1 of the alert macaque monkey.
Responses from neurons in area V1 of the alert macaque monkey to textured patterns modeled after stimuli used in psychophysical experiments of pop- out are consistent with a possible functional role of V1 cells in the mediation of perceptual pop-out and in the segregation of texture borders.
Task-Driven Convolutional Recurrent Models of the Visual System
It is found that standard forms of recurrence do not perform well within deep CNNs on the ImageNet task, and novel cells that incorporated two structural features were able to boost task accuracy substantially, suggesting a role for the brain's recurrent connections in performing difficult visual behaviors.