| 講演抄録／キーワード |
| 講演名 |
2020-02-28 16:10
Compact model mapping based on precise sparse point cloud matching
○HungYa Tsai・Yuya Ieiri・Reiko Hishiyama（Waseda University） |
| 抄録 |
（和） |
Augmented Reality (AR) has gained a lot of attention in recent years due to its limitless imagination. Model
mapping is one of the most important factors that make the AR more realistic and immersive. Previous works require tons of images to reconstruct the dense 3D point cloud for further matching, however, it takes lots of time and cost. In this work, we present a sparse point cloud generating algorithm, which assigns a weight to each captured point cloud based on its distinctiveness and distribution. We retain the most distinctive points for further matching, projecting and mapping the virtual model onto the physical one. Experiments show that this approach can save a considerable amount of time while maintaining similar matching and mapping accuracy |
| （英） |
Augmented Reality (AR) has gained a lot of attention in recent years due to its limitless imagination. Model
mapping is one of the most important factors that make the AR more realistic and immersive. Previous works require tons of images to reconstruct the dense 3D point cloud for further matching, however, it takes lots of time and cost. In this work, we present a sparse point cloud generating algorithm, which assigns a weight to each captured point cloud based on its distinctiveness and distribution. We retain the most distinctive points for further matching, projecting and mapping the virtual model onto the physical one. Experiments show that this approach can save a considerable amount of time while maintaining similar matching and mapping accuracy |
| キーワード |
（和） |
Augmented Reality / Model Mapping / Point Cloud / Photogrammetry / / / / |
| （英） |
Augmented Reality / Model Mapping / Point Cloud / Photogrammetry / / / / |
| 文献情報 |
映情学技報 |
| 資料番号 |
|
| 発行日 |
|
| ISSN |
|
| PDFダウンロード |
|
|