Geometry-Based Skin Colour Estimation for Bare Torso Surface Reconstruction

• Tipo: Artigo em Livro de Atas de Conferência Internacional
• Data: 2019

Autores

Monteiro, JP (Autor) Outra; Zolfagharnasab, H (Autor) Outra; Oliveira, HP (Autor) FCUP

Ata de Conferência Internacional

Título: IMAGE ANALYSIS AND PROCESSING - ICIAP 2019, PT II

Páginas: 595-606

20th International Conference on Image Analysis and Processing (ICIAP)

Univ Trento, Fac Law, Trento, ITALY, SEP 09-13, 2019

Indexação

ISI Web of Knowledge - 0 Citações

Scopus - 0 Citações

Outras Informações

ID Authenticus: P-00R-4FZ

DOI: 10.1007/978-3-030-30645-8_54

Abstract (EN): Three-dimensional imaging techniques have been endeavouring at reaching affordable ubiquity. Nevertheless, its use in clinical practice can be hampered by less than naturally looking surfaces that greatly impact its visual inspection. This work considers the task of surface reconstruction from point clouds of non-rigid scenes acquired through structured-light-based methods, wherein the reconstructed surface contains some level of imperfection to be inpainted before visualized by experts in a clinically oriented context. Appertain to the topic, the recovery of colour information for missing or damaged partial regions is considered. A local geometry-based interpolation method is proposed for the reconstruction of the bare human torso and compared against a reference differential equations based inpainting method. Widely used perceptual distance-based metrics, such as PSNR, SSIM and MS-SSIM, and the evaluation from a panel of experienced breast cancer surgeons is presented for the discussion on inpainting quality assessment.

Idioma: Inglês

Tipo (Avaliação Docente): Científica

Nº de páginas: 12