タイトル(掲載誌)Frontiers of Computer Vision
一般注記In this paper, we discuss a 3D shape reconstruction method of Japanese traditional puppet heads for a digital archiving. Especially, to reconstruct an inner shape of head, we use CT images. First, we divide four regions (wood, hair, paint, and air) by thresholds based on manual directed regions. After that, we divide these regions by a graph cut method. And we also present a method to estimate 3D shape of parts in puppet head. This method is also based on a graph cut method. Moreover, we also discuss a method to distinguish material of puppet head by machine learning. Here we use “U-Net” to extract wood parts of puppet head from its CT images. And we show experimental results by these methods.
一次資料へのリンクURLhttp://repo.lib.tokushima-u.ac.jp/files/public/11/118059/20230313113605108473/fcv2022_49.pdf
関連情報(DOI)10.1007/978-3-031-06381-7_4
連携機関・データベース国立情報学研究所 : 学術機関リポジトリデータベース(IRDB)(機関リポジトリ)