Contrast Sensitivity Functions in Autoencoders

Qiang Li, Alex Gomez-Villa, Marcelo Bertalmio and Jesus Malo

Journal of Vision (Special issue on Deep Nets in Biological Vision) 2021


Three decades ago, Atick et al. suggested that human frequency sensitivity may emerge from the enhancement required for a more efficient analysis of retinal images. Here we reassess the relevance of low-level vision tasks in the explanation of the Contrast Sensitivity Functions (CSFs) in light of (1) the current trend of using artificial neural networks for studying vision, and (2) the current knowledge of retinal image representations.
As a first contribution, we show that a very popular type of convolutional neural networks (CNNs), called autoencoders, may develop human-like CSFs in the spatio-temporal and chromatic dimensions when trained to perform some basic low-level vision tasks (like retinal noise and optical blur removal), but not others (like chromatic adaptation or pure reconstruction after simple bottlenecks). As an illustrative example, the best CNN (in the considered set of simple architectures for enhancement of the retinal signal) reproduces the CSFs with an RMSE error of 11% of the maximum sensitivity.
As a second contribution, we provide experimental evidence of the fact that, for some functional goals (at low abstraction level), deeper CNNs that are better in reaching the quantitative goal are actually worse in replicating human-like phenomena (such as the CSFs). This low-level result (for the explored networks) is not necessarily in contradiction with other works that report advantages of deeper nets in modeling higher-level vision goals. However, in line with a growing body of literature, our results suggests another word of caution about CNNs in vision science since the use of simplified units or unrealistic architectures in goal optimization may be a limitation for the modeling and understanding of human vision.





Back to top

Download Data and Code!



Back to top

Citation and References

PAPER:

Q.Li, A. Gómez-Villa, M. Bertalmío and J. Malo.
Contrast Sensitivity Functions in Autoencoders
Accepted in J. Vision (Special Issue on Deep Nets in Biological Vision) ArXiV: Quantit. Biol. (2021)

https://arxiv.org/abs/2103.00481 mat

Conversation with reviewers in the Journal of Vision:

J. Malo and Q. Li,
Submitted in Feb. 2021, accepted Dec. 2021

Project Notebook:

J. Malo and Q. Li,
CSF in autoencoders. Notebook (as of August 2021)
Evolving Google Notebook

Related papers:

A. Hepburn, V. Laparra, R. Santos, J. Ballé and J. Malo,
On the relation between statistical learning and perceptual distances. ICLR 2022 (see OpenReview)
ArXiV: Comp. Vis. Patt. Recogn. (2021) https://arxiv.org/abs/2106.04427 mat Arxiv Preprint