ORIGINAL RESEARCH article
Front. Artif. Intell.
Sec. Machine Learning and Artificial Intelligence
A Turing Test for Artificial Nets devoted to Vision
Provisionally accepted
- 1Image Processing Lab, Universitat de Valencia, Valencia, Spain
- 2Universitat de Valencia, Valencia, Spain
- 3TReNDs, Georgia Institute of Technology, Atlanta, United States
- 4Image Processing Lab, University of Valencia, Valencia, Spain
Select one of your emails
You have multiple emails registered with Frontiers:
Notify me on publication
Please enter your email address:
If you already have an account, please login
You don't have a Frontiers account ? You can register here
In this work1 we argue that, despite recent claims about successful modeling of the visual brain using deep nets, the problem is far from being solved, particularly for low-level vision. Open issues include where should we read from in ANNs to check behavior?, what should be the read-out?, this ad-hoc read-out is considered part of the brain model or not?, in order to understand vision-ANNs, should we use artificial psychophysics or artificial physiology?, anyhow, artificial tests should literally match the experiments done with humans?. These questions suggest a clear need of biologically sensible tests for deep models of the visual brain, and more generally, to understand ANNs devoted to generic vision tasks. Following our use of low-level facts from Vision Science in Image Processing, we present a low-level dataset compiling the basic spatio-chromatic properties that describe the adaptive bottleneck of the retina-V1 pathway, and are not currently available in popular databases such as BrainScore. We propose its use for qualitative and quantitative model evaluation. As illustration of the proposed methods we check the behavior of three recent models with similar deep architecture: (1) A parametric model tuned via the psychophysical method of Maximum Differentiation [Malo & Simoncelli SPIE 15, Martinez et al. PLOS 18, Martinez et al. Front. Neurosci. 19], (2) A non-parametric model (the PerceptNet) tuned to maximize the correlation with humans on subjective image distortions [Hepburn et al. IEEE ICIP 20], and (3) A model with the same encoder as the PerceptNet, but tuned for image segmentation [Hernandez-Camara et al. Patt.Recogn.Lett. 23, Hernandez-Camara et al. Neurocomp. 25]. Results on the proposed 10 compelling psycho/physio visual properties show that the first (parametric) model is the one with closer behavior to humans.
Keywords: Evaluation of AI models, Neural Networks for Vision, human vision, Turing test, Low-level Visual Psychophysics, Linear+Nonlinear cascade, image quality, image segmentation
Received: 14 Jul 2025; Accepted: 17 Nov 2025.
Copyright: © 2025 Vila-Tomás, Hernández-Cámara, Li, Laparra and Malo. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
* Correspondence: Jesús Malo, jesus.malo@uv.es
Disclaimer: All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article or claim that may be made by its manufacturer is not guaranteed or endorsed by the publisher.