Skip to main content

Bas-relief modelling from enriched detail and geometry with deep normal transfer.

Wang, M., Wang, L., Jiang, T., Xiang, N., Lin, J., Wei, M., Yang, X., Komura, T. and Zhang, J. J., 2020. Bas-relief modelling from enriched detail and geometry with deep normal transfer. Neurocomputing. (In Press)

Full text available as:

[img] PDF
elsarticle-template-final-V3.pdf - Accepted Version
Restricted to Repository staff only until 14 September 2021.
Available under License Creative Commons Attribution Non-commercial.

13MB

DOI: 10.1016/j.neucom.2020.06.130

Abstract

Detail-and-geometry richness is essential to bas-relief modelling. However, existing image-based and model-based bas-relief modelling techniques commonly suffer from detail monotony or geometry loss. In this paper, we introduce a new bas-relief modelling framework for detail abundance with visual attention based mask generation and geometry preservation, which benefits from our two key contributions. For detail richness, we propose a novel semantic neural network of normal transfer to enrich the texture styles on bas-reliefs. For geometry preservation, we introduce a normal decomposition scheme based on Domain Transfer Recursive Filter (DTRF). Experimental results demonstrate that our approach is advantageous on producing bas-relief modellings with both fine details and geometry preservation.

Item Type:Article
ISSN:0925-2312
Uncontrolled Keywords:Bas-relief modelling; Normal transfer; Image-based normal decomposition; Detail transfer; Geometry preservation; Visual attention
Group:Faculty of Media & Communication
ID Code:34845
Deposited By: Unnamed user with email symplectic@symplectic
Deposited On:18 Nov 2020 14:58
Last Modified:18 Nov 2020 14:58

Downloads

Downloads per month over past year

More statistics for this item...
Repository Staff Only -