Neural Networks Trained on Natural Scenes Exhibit Gestalt Closure
This addresses the problem of understanding the origins of perceptual grouping laws in vision science, offering an incremental insight by linking deep learning to classical psychological theories.
The study investigated whether natural scene statistics can derive the Gestalt law of closure, and found that a convolutional neural network trained on natural images exhibits closure on synthetic edge fragments, supporting the hypothesis that adaptation to environmental statistics might suffice as a fundamental perceptual principle.
The Gestalt laws of perceptual organization, which describe how visual elements in an image are grouped and interpreted, have traditionally been thought of as innate despite their ecological validity. We use deep-learning methods to investigate whether natural scene statistics might be sufficient to derive the Gestalt laws. We examine the law of closure, which asserts that human visual perception tends to "close the gap" by assembling elements that can jointly be interpreted as a complete figure or object. We demonstrate that a state-of-the-art convolutional neural network, trained to classify natural images, exhibits closure on synthetic displays of edge fragments, as assessed by similarity of internal representations. This finding provides support for the hypothesis that the human perceptual system is even more elegant than the Gestaltists imagined: a single law---adaptation to the statistical structure of the environment---might suffice as fundamental.