ETRI-Knowledge Sharing Plaform

KOREAN
논문 검색
Type SCI
Year ~ Keyword

Detail

Conference Paper Simulation of 3D Virtual Space Model Based on Texture Image and Structure Analysis
Cited 2 time in scopus Share share facebook twitter linkedin kakaostory
Authors
Dongwan Ryoo, Kyungkyu Kang, Changjoon Park
Issue Date
2019-10
Citation
International Conference on Information and Communication Technology Convergence (ICTC) 2019, pp.1321-1323
Publisher
IEEE
Language
English
Type
Conference Paper
DOI
https://dx.doi.org/10.1109/ICTC46691.2019.8939767
Abstract
In this paper, we employ a 3D reconstruction technique using the multi-view images, i.e., one of the 3D data generation methods. we provide an algorithm for simulating seasonal changes, such as autumn or winter contents, from summer contents in a 3D virtual space model. This algorithm uses two-dimensional (2D) texture analysis and 3D structure analysis in the virtual space model for the simulation. The 2D texture analysis is based on the color and edge components of the texture image. The 3D structure analysis is performed using the normal vector components corresponding to the mesh of the objects. The advantages of the proposed algorithm are as follows. It is efficient in terms of time and cost, as it is not necessary to take pictures every season for producing seasonal 3D contents. This algorithm provides increased efficiency, owing to the small processing computations. It is possible to provide various contents to the user by using only one input source.
KSP Keywords
3D Reconstruction, 3D content, 3D data, 3D structure analysis, Data generation, Reconstruction technique, Texture image, Virtual space, input source, model-based, multi-view images