M²LIC: An Automated Target-Free Mobile Pipeline for Creating Relightable Models

Ruggero Pintus, Antonio Zorcolo, Alberto Jaspe-Villanueva, Nicola Pagani, Pietro Francesco Serreli, Giovanna Pietra, Enrico Gobbetti

Eurographics Workshop on Graphics and Cultural Heritage, GCH'26 RelightingCultural heritageCapture and reconstruction

M²LIC: An Automated Target-Free Mobile Pipeline for Creating Relightable Models teaser

PDF DOI EG Digital Library Journal version (JOCCH)

@InProceedings{Pintus:2026:M2LICShort,
 title = {{M$^2$LIC}: An Automated Target-Free Mobile Pipeline for Creating Relightable Models},
 author = {Pintus, Ruggero and Zorcolo, Antonio and Jaspe-Villanueva, Alberto and Pagani, Nicola and Serreli, Pietro Francesco and Pietra, Giovanna and Gobbetti, Enrico},
 booktitle = {Eurographics Workshop on Graphics and Cultural Heritage},
 editor = {Pintus, Ruggero and Bosch, Carles and Schroer, Carla and Charalambous, Panagiotis and Ciortan, Irina and Kotzinos, Dimitris and Andujar, Carlos and Dininno, Domenica},
 publisher = {The Eurographics Association},
 month = {November},
 year = {2026},
 isbn = {978-3-03868-329-2},
 issn = {2312-6124},
 doi = {10.2312/gch.20261061}
 }

Abstract

Multi-Light Image Collections (MLIC) are fundamental for virtual re-lighting and surface analysis in cultural heritage. However, most standard free-form acquisition remains hindered by the need for calibration targets, specialized training, and manual processing. We present M²LIC (Mobile Multi-Light Image Collection), a fully automated, target-free pipeline utilizing just two commodity smartphones, i.e., one fixed for capture, one moving for illumination. Our framework synchronizes video streams, performs frame selection to identify a small set of sharp and well-distributed images, infers light directions directly from the scene structure, and automatically generates web-ready relightable models. Validated on diverse artifacts, this rapid, open-source protocol significantly lowers the barrier to entry, enabling highly scalable MLIC digitization workflows.