# Paraphrasing Magritte's Observation

arXiv: https://arxiv.org/abs/2202.08103 · HTML: https://mlsanity.com/paper/2202.08103/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 6 of its 15 references that are in the map.
- First posted on arXiv 11 Feb 2022.

## Builds on (most cited first, 6 of 6)

- [ImageNet Large Scale Visual Recognition Challenge](https://mlsanity.com/paper/1409.0575/) (2014)
- [Contrast Sensitivity Functions in Autoencoders](https://mlsanity.com/paper/2103.00481/) (2021)
- [Nonlinearities and Adaptation of Color Vision from Sequential Principal Curves Analysis](https://mlsanity.com/paper/1602.00236/) (2016)
- [Derivatives and Inverse of Cascaded Linear+Nonlinear Neural Models](https://mlsanity.com/paper/1711.00526/) (2017)
- [Spatio-Chromatic Information available from different Neural Layers via Gaussianization](https://mlsanity.com/paper/1910.01559/) (2019)
- [Visual Information flow in Wilson-Cowan networks](https://mlsanity.com/paper/1907.13046/) (2019)

## Abstract (arXiv)

Contrast Sensitivity of the human visual system can be explained from certain low-level vision tasks (like retinal noise and optical blur removal), but not from others (like chromatic adaptation or pure reconstruction after simple bottlenecks). This conclusion still holds even under substantial change in stimulus statistics, as for instance considering cartoon-like images as opposed to natural images (Li et al. Journal of Vision, 2022, Preprint arXiv:2103.00481). In this note we present a method to generate original cartoon-like images compatible with the statistical training used in (Li et al., 2022). Following the classical observation in (Magritte, 1929), the stimuli generated by the proposed method certainly are not what they represent: Ceci n'est pas une pipe. The clear distinction between representation (the stimuli generated by the proposed method) and reality (the actual object) avoids eventual problems for the use of the generated stimuli in academic, non-profit, publications.