スタイル変換に基づくデータ拡張を用いた三次元物体復元手法

カテゴリ: 論文誌(論文単位)

グループ名: 【C】電子・情報・システム部門

発行日: 2020/11/01

タイトル(英語): 3D Reconstruction Based on Style Transfer Data Augmentation

著者名: 猿渡　太貴（大阪府立大学 大学院工学研究科），井上　勝文（大阪府立大学 大学院工学研究科），吉岡　理文（大阪府立大学 大学院工学研究科）

著者名(英語): Taiki Saruwatari (Graduate School of Engineering, Osaka Prefecture University), Katsufumi Inoue (Graduate School of Engineering, Osaka Prefecture University), Michifumi Yoshioka (Graduate School of Engineering, Osaka Prefecture University)

キーワード: 三次元復元，スタイル変換，三次元点群　　3D reconstruction，style transfer，point cloud

要約(英語): Recently, various technologies using PointCloud and Deep Neural Network (DNN) have been actively researched. However, there is a disadvantage that the collecting PointCloud data from real object with special sensors such as depth sensor is time consuming task. To deal with this problem, we focus on 3D reconstruction from a single image. Conventional methods construct PointCloud from a single image which includes mask information. Therefore, it is difficult to construct a PointCloud from an image without mask information. To remove the requirement of the additional information such as mask for input image, we propose data augmentation based on style transfer for 3D reconstruction. It is known that DNN using style transformed image can learn a shape feature. By using the transformed images, the DNN can learn object shapes with various backgrounds and textures and can obtain shape features even from the images with cluttered background. From the experimental results, we confirmed that our proposed method could construct 3D object shape with PointCloud while keeping shape information without additional information.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.140 No.11 （2020） 特集：電気関係学会関西連合大会

本誌掲載ページ: 1198-1206 p

原稿種別: 論文／日本語

電子版へのリンク: https://www.jstage.jst.go.jp/article/ieejeiss/140/11/140_1198/_article/-char/ja/