AI-driven specular removal for 3D asset creation
Callieri M., Corsini M., Dutta S., Giorgi D., Sorrenti M.
Solid modeling Three-dimensional displays Pipelines Neural networks Digital signal processing Sun Image reconstruction Standards Photogrammetry Synthetic data
Specular highlights negatively affect photogram-metric 3D reconstructions. To mitigate this problem, we developed an AI-driven image processing technique able to remove specular highlights. We created a synthetic image dataset that reflects the objects, viewpoints, and specular behaviors found in real-world photogrammetric campaigns, and used it to train a U-Net model that can batch-process input images for photogrammetric reconstruction. The process was tested on both synthetic and real-world photos, demonstrating superior results compared to existing models in the literature.
Publisher: Institute of Electrical and Electronics Engineers Inc.
Metrics
Back to previous page
@inproceedings{oai:iris.cnr.it:20.500.14243/552645,
title = {AI-driven specular removal for 3D asset creation},
author = {Callieri M. and Corsini M. and Dutta S. and Giorgi D. and Sorrenti M.},
publisher = {Institute of Electrical and Electronics Engineers Inc.},
doi = {10.1109/dsp65409.2025.11075117},
year = {2025}
}
0000-0002-5832-680XBibliographic record
Deposited version
Deposited version
10.1109/dsp65409.2025.11075117
