Material recognition from cross-category morphing
Abstract
Materials from distinct categories share common image features but also exhibit unique characteristics. How do humans recognize materials despite their enormous appearance variations? Using generative neural networks, we create both prototypical and ambiguous materials that morph between categories. To capture the richness of material representation, we characterize how people judge materials using cross-material morphs through three behavioral tasks: material categorization, material property rating, and visual similarity judgment. We find that morphing smoothly modulates perceptual scales of material appearance. Despite this smooth variation, participants can reliably identify the prototypical materials within given appearance ranges, indicating a strong association between material categories and their visual characteristics. Material properties structure the perceptual space, where the salient dimensions–particularly rigidity–strongly correlate with perceived material categories. In contrast, analyzing image embedding derived from a self-supervised deep learning model reveals that learning visual similarities alone is insufficient to reproduce human perceptual space of materials. Together, our results suggest that material representation is more than visual similarities and may require learning material properties that structure relationships among materials within and across categories. Visual features might be used flexibly and task-dependently to support multiple levels of material perception, ranging from discrimination to material property inference to categorization.
Related
- The ViRA-L (Visual Recognition Assessment-Leuven): Development, Validation, and Age Stratified Normative Characterization of a Multi-Category Visual Recognition Battery
- Differential Effects of Task-Context and Stimulus-Category Changes on Recognition Memory: A Replication and Extension of Morse et al. (2023)
- Associations of the metabolic score for insulin resistance and uric acid with prevalent thyroid nodules: A cross-sectional study with systolic blood pressure mediation analysis and C-TIRADS category-specific analyses
- Smallpox Vaccine Elicits CD8+ T Cell Immunity Persisting Over Four Decades with Cross-Recognition of Monkeypox Virus
- Long-term impact of the diagnosis on quality of life, social skills, and person recognition strategies in congenital prosopagnosics: A cross-sectional study
BibTeX
@article{liao2026material,
title = {{Material recognition from cross-category morphing}},
author = {Chenxi Liao and Filipp Schmidt and Jacob R. Cheeseman and Masataka Sawayama and Roland W. Fleming and Bei Xiao},
journal = {Scientific Reports},
year = {2026},
month = oct,
doi = {10.1038/s41598-026-59920-x},
url = {https://www.nature.com/articles/s41598-026-59920-x},
note = {CC BY 4.0},
}
RIS
TY - JOUR
TI - Material recognition from cross-category morphing
AU - Chenxi Liao
AU - Filipp Schmidt
AU - Jacob R. Cheeseman
AU - Masataka Sawayama
AU - Roland W. Fleming
AU - Bei Xiao
PY - 2026
DA - 2026/10/01
JO - Scientific Reports
DO - 10.1038/s41598-026-59920-x
UR - https://www.nature.com/articles/s41598-026-59920-x
AB - Materials from distinct categories share common image features but also exhibit unique characteristics. How do humans recognize materials despite their enormous appearance variations? Using generative neural networks, we create both prototypical and ambiguous materials that morph between categories. To capture the richness of material representation, we characterize how people judge materials using cross-material morphs through three behavioral tasks: material categorization, material property rating, and visual similarity judgment. We find that morphing smoothly modulates perceptual scales of material appearance. Despite this smooth variation, participants can reliably identify the prototypical materials within given appearance ranges, indicating a strong association between material categories and their visual characteristics. Material properties structure the perceptual space, where the salient dimensions–particularly rigidity–strongly correlate with perceived material categories. In contrast, analyzing image embedding derived from a self-supervised deep learning model reveals that learning visual similarities alone is insufficient to reproduce human perceptual space of materials. Together, our results suggest that material representation is more than visual similarities and may require learning material properties that structure relationships among materials within and across categories. Visual features might be used flexibly and task-dependently to support multiple levels of material perception, ranging from discrimination to material property inference to categorization.
SN - 2045-2322
ER -