Evaluating CNNs on the Gestalt Principle of Closure

03/30/2019

∙

Deep convolutional neural networks (CNNs) are widely known for their outstanding performance in classification and regression tasks over high-dimensional data. This made them a popular and powerful tool for a large variety of applications in industry and academia. Recent publications show that seemingly easy classifaction tasks (for humans) can be very challenging for state of the art CNNs. An attempt to describe how humans perceive visual elements is given by the Gestalt principles. In this paper we evaluate AlexNet and GoogLeNet regarding their performance on classifying the correctness of the well known Kanizsa triangles, which heavily rely on the Gestalt principle of closure. Therefore we created various datasets containing valid as well as invalid variants of the Kanizsa triangle. Our findings suggest that perceiving objects by utilizing the principle of closure is very challenging for the applied network architectures but they appear to adapt to the effect of closure.

READ FULL TEXT

Evaluating CNNs on the Gestalt Principle of Closure

Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure

Evaluation of deep convolutional neural networks in classifying human embryo images based on their morphological quality

Can CNNs Construct Highly Accurate Model Efficiently with Limited Training Samples?

Representations for the largest Extension of a closure system

CP-CNN: Core-Periphery Principle Guided Convolutional Neural Network

What Deep CNNs Benefit from Global Covariance Pooling: An Optimization Perspective

A neural network closure for the Euler-Poisson system based on kinetic simulations

Evaluating CNNs on the Gestalt Principle of Closure

Related Research

Do Neural Networks Show Gestalt Phenomena? An Exploration of the Law of Closure

Evaluation of deep convolutional neural networks in classifying human embryo images based on their morphological quality

Can CNNs Construct Highly Accurate Model Efficiently with Limited Training Samples?

Representations for the largest Extension of a closure system

CP-CNN: Core-Periphery Principle Guided Convolutional Neural Network

What Deep CNNs Benefit from Global Covariance Pooling: An Optimization Perspective

A neural network closure for the Euler-Poisson system based on kinetic simulations