Dynamic decorrelation as a unifying principle for explaining a broad range of brightness phenomena

Abstract

The visual system is highly sensitive to spatial context for encoding luminance patterns. Context sensitivity inspired the proposal of many neural mechanisms for explaining the perception of luminance (brightness). Here we propose a novel computational model for estimating the brightness of many visual illusions. We hypothesize that many aspects of brightness can be explained by a dynamic filtering process that reduces the redundancy in edge representations on the one hand, while non-redundant activity is enhanced on the other. The dynamic filter is learned for each input image and implements context sensitivity. Dynamic filtering is applied to the responses of (model) complex cells in order to build a gain control map. The gain control map then acts on simple cell responses before they are used to create a brightness map via activity propagation. Our approach is successful in predicting many challenging visual illusions, including contrast effects, assimilation, and reverse contrast with the same set of model parameters.

Simple Assumptions to Improve Markov Illuminance and Reflectance

“…White's effect has long been known as a problem for simple contrast-based theories and has been discussed by many lightness/brightness researchers (Spehar et al, 1995;Taya et al, 1995;Anderson et al, 2001;Howe, 2005;Gilchrist, 2006;Betz et al, 2015). Although there have been various qualitative explanations of White's effect (e.g., Howe, 2001), not many quantitative (i.e., computational) models have correctly explained it (see Blakeslee and McCourt, 1999;Lerer et al, 2021 for successful cases). Therefore, it is highly important for computational models to incorporate the prediction of White's effect.…”

Synthetic surprise as the foundation of the psychedelic experience

“…It posits that perception is an active process, where top-down predictions shape and guide the processing of bottom-up sensory inputs. This framework can account for various perceptual phenomena, including perceptual illusions and the influence of prior knowledge on perception ( Lerer et al, 2021 , Weilnhammer et al, 2017 , Lumer et al, 1998 ).…”

Empirical evaluation of computational models of lightness perception

“…Dynamic decorrelation : Lerer et al 28 proposed a model of brightness perception based on the principle of dynamic decorrelation. They suggested that brightness perception is a consequence of suppressing redundant, that is, predictable information.…”

“…This is achieved by a dynamic filter that adapts to the spatial structure of luminance patterns in the input image. Lerer et al 28 showed that their model is able to correctly predict brightness perception in a large set of illusions and phenomena. It also successfully handles real-world images and images corrupted by noise.…”