2021 Fiscal Year Final Research Report
Fundamental Optical Factorization of Monocular Images
| Project/Area Number |
19H04130
|
|---|
| Research Category |
Grant-in-Aid for Scientific Research (B)
|
| Allocation Type | Single-year Grants |
|---|
| Section | 一般 |
|---|
| Review Section |
Basic Section 61010:Perceptual information processing-related
|
|---|
| Research Institution | University of Tsukuba |
|---|
Principal Investigator |
|
|---|
| Co-Investigator(Kenkyū-buntansha) |
遠藤 結城 筑波大学, システム情報系, 助教 (00790396)
|
|---|
| Project Period (FY) |
2019-04-01 – 2022-03-31
|
| Keywords | 画像分解 / 深層学習 / 光源・形状・反射率推定 / 単眼三次元復元 / 映像編集 |
|---|
| Outline of Final Research Achievements |
Given objects in monocular input images, we conducted research projects toward various applications based on optical factorization of target objects into objects’ color/texture, shape, and illumination. We obtained the following research outcomes. We published technical papers on 1) semantic segmentation of human 3D models (domestic meeting: two), 2) optical factorization for full-body human images (invited talk: one, invited article: one, domestic meeting: one, journal/international conference: one), and 3) factorization and 3D reconstruction of terrain images (domestic meeting: two, journal submission: one). Also, we acquired a patent on optical factorization.
|
| Free Research Field |
コンピュータグラフィックス、コンピュータビジョン
|
| Academic Significance and Societal Importance of the Research Achievements |
光学要素分解とは、被写体の色・模様、形状、被写体を照らす光源に由来する成分へ入力画像を分解する基盤技術であり、物体認識、反射率推定、形状推定、光源推定など、多数の応用先がある。本研究の成果は、画像・映像編集、照明設計、単視点映像から物体認識を行うロボット制御やヒューマンコンピュータインタラクション、三次元モデルを必要とするAR/MR/VRなど、広範な分野で活用できる。
|
