From Noisy Point Clouds to Complete Ear Shapes: Unsupervised Pipeline
Published in IEEE Access, 2021
Recommended citation: Filipa Valdeira, Ricardo Ferreira, Alessandra Micheletti, Claudia Soares, "From Noisy Point Clouds to Complete Ear Shapes: Unsupervised Pipeline." IEEE Access, 2021. http://dx.doi.org/10.1109/access.2021.3111811
Abstract
Ears are a particularly difficult region of the human face to model, not only due to the non-rigid deformations existing between shapes but also to the challenges in processing the retrieved data. The first step towards obtaining a good model is to have complete scans in correspondence, but these usually present a higher amount of occlusions, noise and outliers when compared to most face regions, thus requiring a specific procedure. Therefore, we propose a complete pipeline taking as input unordered 3D point clouds with the aforementioned problems, and producing as output a dataset in correspondence, with completion of the missing data. We provide a comparison of several state-of-the-art registration and shape completion methods, concluding on the best choice for each of the steps.
BibTeX
@article{Valdeira_2021,
author = "Valdeira, Filipa and Ferreira, Ricardo and Micheletti, Alessandra and Soares, Claudia",
title = "From Noisy Point Clouds to Complete Ear Shapes: Unsupervised Pipeline",
volume = "9",
ISSN = "2169-3536",
url = "http://dx.doi.org/10.1109/access.2021.3111811",
DOI = "10.1109/access.2021.3111811",
journal = "IEEE Access",
publisher = "Institute of Electrical and Electronics Engineers (IEEE)",
year = "2021",
pages = "127720-127734",
abstract = "Ears are a particularly difficult region of the human face to model, not only due to the non-rigid deformations existing between shapes but also to the challenges in processing the retrieved data. The first step towards obtaining a good model is to have complete scans in correspondence, but these usually present a higher amount of occlusions, noise and outliers when compared to most face regions, thus requiring a specific procedure. Therefore, we propose a complete pipeline taking as input unordered 3D point clouds with the aforementioned problems, and producing as output a dataset in correspondence, with completion of the missing data. We provide a comparison of several state-of-the-art registration and shape completion methods, concluding on the best choice for each of the steps.",
abstractsource = "https://run.unl.pt/server/api/core/bitstreams/3e4f2a64-13d7-458a-84d3-48c0a37bff0c/content",
abstractsourcelabel = "Published abstract source"
}