A 3D Reconstruction Method with Color Reproduction from Multi-band and Multi-view Images
| Autor: | Takafumi Aoki, Koichi Ito, Shuya Ito, Masaru Tsuchida |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
Color histogram
Computer science business.industry 3D reconstruction ComputingMethodologies_IMAGEPROCESSINGANDCOMPUTERVISION Point cloud Reprojection error Color reproduction 020207 software engineering 02 engineering and technology Rendering (computer graphics) Mesh generation 0202 electrical engineering electronic engineering information engineering Structure from motion 020201 artificial intelligence & image processing Computer vision Artificial intelligence business |
| Zdroj: | Computer Vision – ACCV 2016 Workshops ISBN: 9783319544267 ACCV Workshops (2) |
| DOI: | 10.1007/978-3-319-54427-4_18 |
| Popis: | The purpose of digital archiving is the accurate rendering of authenticated contents in terms of color and shape. This paper proposes a novel 3D reconstruction method with color reproduction from multi-band and multi-view images. The proposed method consists of 5 steps: (i) feature tracking of multi-band and multi-view images, (ii) sparse 3D reconstruction using Structure from Motion (SfM), (iii) mesh generation from the sparse 3D point clouds, (iv) dense 3D reconstruction and (v) color reproduction. We create multi-band and multi-view image datasets of Japanese dolls with Kimono for quantitative performance evaluation. Through a set of experiments using our datasets, we confirm that our proposed method exhibits efficient performance of reconstructing a high-quality 3D model from multi-band and multi-view images compared with conventional methods. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|