Leveraging Geometric Visual Illusions as Perceptual Inductive Biases for Vision Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Yang, Haobo, Guo, Minghao, Yang, Dequan, Wang, Wenyu
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916961341931520
author Yang, Haobo
Guo, Minghao
Yang, Dequan
Wang, Wenyu
author_facet Yang, Haobo
Guo, Minghao
Yang, Dequan
Wang, Wenyu
contents Contemporary deep learning models have achieved impressive performance in image classification by primarily leveraging statistical regularities within large datasets, but they rarely incorporate structured insights drawn directly from perceptual psychology. To explore the potential of perceptually motivated inductive biases, we propose integrating classic geometric visual illusions well-studied phenomena from human perception into standard image-classification training pipelines. Specifically, we introduce a synthetic, parametric geometric-illusion dataset and evaluate three multi-source learning strategies that combine illusion recognition tasks with ImageNet classification objectives. Our experiments reveal two key conceptual insights: (i) incorporating geometric illusions as auxiliary supervision systematically improves generalization, especially in visually challenging cases involving intricate contours and fine textures; and (ii) perceptually driven inductive biases, even when derived from synthetic stimuli traditionally considered unrelated to natural image recognition, can enhance the structural sensitivity of both CNN and transformer-based architectures. These results demonstrate a novel integration of perceptual science and machine learning and suggest new directions for embedding perceptual priors into vision model design.
format Preprint
id arxiv_https___arxiv_org_abs_2509_15156
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Leveraging Geometric Visual Illusions as Perceptual Inductive Biases for Vision Models
Yang, Haobo
Guo, Minghao
Yang, Dequan
Wang, Wenyu
Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
Contemporary deep learning models have achieved impressive performance in image classification by primarily leveraging statistical regularities within large datasets, but they rarely incorporate structured insights drawn directly from perceptual psychology. To explore the potential of perceptually motivated inductive biases, we propose integrating classic geometric visual illusions well-studied phenomena from human perception into standard image-classification training pipelines. Specifically, we introduce a synthetic, parametric geometric-illusion dataset and evaluate three multi-source learning strategies that combine illusion recognition tasks with ImageNet classification objectives. Our experiments reveal two key conceptual insights: (i) incorporating geometric illusions as auxiliary supervision systematically improves generalization, especially in visually challenging cases involving intricate contours and fine textures; and (ii) perceptually driven inductive biases, even when derived from synthetic stimuli traditionally considered unrelated to natural image recognition, can enhance the structural sensitivity of both CNN and transformer-based architectures. These results demonstrate a novel integration of perceptual science and machine learning and suggest new directions for embedding perceptual priors into vision model design.
title Leveraging Geometric Visual Illusions as Perceptual Inductive Biases for Vision Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Machine Learning
url https://arxiv.org/abs/2509.15156