Unsupervised Learning Predicts Human Perception and Misperception of Specular Surface Reflectance

BioRxiv : the Preprint Server for Biology
Katherine R Storrs, Roland W Fleming


Gloss perception is a challenging visual inference that requires disentangling the contributions of reflectance, lighting, and shape to the retinal image. Learning to see gloss must somehow proceed without labelled training data as no other sensory signals can provide the 'ground truth' required for supervised learning. We reasoned that paradoxically, we may learn to infer distal scene properties, like gloss, by learning to compress and predict spatial structure in proximal image data. We hypothesised that such unsupervised learning might explain both successes and failures of human gloss perception, where classical 'inverse optics' cannot. To test this, we trained unsupervised neural networks to model the pixel statistics of renderings of glossy surfaces and compared the resulting representations with human gloss judgments. The trained networks spontaneously cluster images according to underlying scene properties such as specular reflectance, shape and illumination, despite receiving no explicit information about them. More importantly, we find that linearly decoding specular reflectance from the model's internal code predicts human perception and misperception of glossiness on an image-by-image basis better than the true phys...Continue Reading

Related Concepts

Positioning Attribute
Genetic Inheritance
Disease Transmission
Population Group

Related Feeds

BioRxiv & MedRxiv Preprints

BioRxiv and MedRxiv are the preprint servers for biology and health sciences respectively, operated by Cold Spring Harbor Laboratory. Here are the latest preprint articles (which are not peer-reviewed) from BioRxiv and MedRxiv.

Related Papers

Journal of Vision
Peter VangorpRoland W Fleming
Vision Research
A C Chadwick, R W Kentridge
The Journal of Pharmacy and Pharmacology
R C Rowe
© 2020 Meta ULC. All rights reserved