Convolutional Neural Networks Deceived by Visual Illusions
This work addresses the problem of aligning CNN behavior with human perception for researchers in computer vision and neuroscience, though it is incremental as it builds on existing CNN models and tasks.
The paper investigates whether CNNs trained for low-level visual tasks like image denoising, deblurring, and color constancy are deceived by visual illusions, finding that they can replicate human responses to these illusions, with performance varying by architecture and spatial pattern size.
Visual illusions teach us that what we see is not always what it is represented in the physical world. Its special nature make them a fascinating tool to test and validate any new vision model proposed. In general, current vision models are based on the concatenation of linear convolutions and non-linear operations. In this paper we get inspiration from the similarity of this structure with the operations present in Convolutional Neural Networks (CNNs). This motivated us to study if CNNs trained for low-level visual tasks are deceived by visual illusions. In particular, we show that CNNs trained for image denoising, image deblurring, and computational color constancy are able to replicate the human response to visual illusions, and that the extent of this replication varies with respect to variation in architecture and spatial pattern size. We believe that this CNNs behaviour appears as a by-product of the training for the low level vision tasks of denoising, color constancy or deblurring. Our work opens a new bridge between human perception and CNNs: in order to obtain CNNs that better replicate human behaviour, we may need to start aiming for them to better replicate visual illusions.