敵対的学習を導入した教師なし学習に基づくオプティカルフロー推定手法

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: 2022/06/01

タイトル(英語): UnFlowGAN: Unsupervised Learning of Optical Flow with Adversarial Learning

著者名: 山﨑　海門（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），井上　勝文（大阪府立大学大学院工学研究科）

著者名(英語): Kaito Yamasaki (Osaka Prefecture University), Michifumi Yoshioka (Osaka Prefecture University), Katsufumi Inoue (Osaka Prefecture University)

キーワード: ニューラルネットワーク，オプティカルフロー，教師なし学習，Adversarial Learning　　neural network，optical flow，unsupervised learning，adversarial learning

要約(英語): Recently, unsupervised learning-based approaches for optical flow estimation have been actively researched. In unsupervised settings, the difference between the input image and the reconstructed image created from the estimated optical flow is minimized to learn the optical flow. Conventional learning methods mainly treated the difference as the pixel-by-pixel brightness error, which might lead to decreasing accuracy of the optical flow because the learning-strategy cannot take the textures into account sufficiently. To deal with the problem, in addition to the brightness error, we propose the introduction of adversarial learning into the evaluation of the input image and the reconstructed image. Our main contribution is that we develop a learning-strategy to capture the textures and the proposed method outperforms the conventional methods on the KITTI benchmarks.

本誌: 電気学会論文誌C（電子・情報・システム部門誌） Vol.142 No.6 （2022）

本誌掲載ページ: 650-659 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/142/6/142_650/_article/-char/ja/