Abstract

Selection is the first step in many image editing processes, enabling faster and simpler modifications of all pixels sharing a common modality. In this work, we present a method for material selection in images, robust to lighting and reflectance variations, which can be used for downstream editing tasks. We rely on vision transformer (ViT) models and leverage their features for selection, proposing a multi-resolution processing strategy that yields finer and more stable selection results than prior methods. Furthermore, we enable selection at two levels: texture and subtexture, leveraging a new two-level material selection (DuMaS) dataset which includes dense annotations for over 800,000 synthetic images, both on the texture and subtexture levels.

Paper

Paper: PDF
Supplemental Document: PDF

Code & Data

Download our DuMaS dataset, as well as our Two-Level Test dataset.
Code coming soon...

Bibtex

@article{guerrero2025matselection, title={{Fine-Grained Spatially Varying Material Selection in Images}}, author={J. {Guerrero-Viu} and M. {Fischer} and I. {Georgiev} and E. {Garces} and D. {Gutierrez} and B. {Masia} and V. {Deschaintre}}, journal = {ACM Transactions on Graphics (Proc. SIGGRAPH Asia)}, year = {2025} }

Acknowledgments

This work has been partially supported by grant PID2022-141539NB-I00, funded by MICIU/AEI/10.13039/501100011033 and by ERDF, EU. This work has also received funding from the Government of Aragon’s Departamento de Educacion, Ciencia y Universidades through the Reference Research Group “Graphics and Imaging Lab” (ref T34_23R) and through the project “HUMAN-VR: Development of a Computational Model for Virtual Reality Perception” (PROY_T25_24). Julia Guerrero-Viu developed part of this work during an Adobe internship, and was also partially supported by the FPU20/02340 predoctoral grant. We thank the members of the Graphics and Imaging Lab, especially Nacho Moreno, Nestor Monzon, and Santiago Jimenez, for insightful discussions, help preparing the figures and final proofreading.