Constant velocity 3D convolution

Yusuke Sekikawa, Kohta Ishikawa, Kosuke Hara, Yuuichi Yoshida, Koichiro Suzuki, Ikuro Sato, Hideo Saito

研究成果: Conference contribution

2 被引用数 (Scopus)

抄録

We propose a novel three-dimensional (3D)-convolution method, cv3dconv, for detecting spatiotemporal features from videos. It reduces the number of sum-of-products of 3D convolution by thousands of times by assuming the constant moving velocity of the camera. We observed that a specific class of video sequences, such as those captured by an in-vehicle camera, can be well approximated with piece-wise linear movements of 2D features in the temporal dimension. Our principal finding is that the 3D kernel, represented by the constant-velocity, can be decomposed into a convolution of a 2D kernel representing the shapes and a 3D kernel representing the velocity. We derived the efficient recursive algorithm for this class of 3D convolution which is exceptionally suited for sparse data, and this parameterized decomposed representation imposes a structured regularization along the temporal direction. We experimentally verified the validity of our approximation using a controlled dataset, and we also showed the effectiveness of cv3dconv for the visual odometry estimation task using real event camera data captured in urban road scene.

本文言語English
ホスト出版物のタイトルProceedings - 2018 International Conference on 3D Vision, 3DV 2018
出版社Institute of Electrical and Electronics Engineers Inc.
ページ343-351
ページ数9
ISBN(電子版)9781538684252
DOI
出版ステータスPublished - 2018 10 12
イベント6th International Conference on 3D Vision, 3DV 2018 - Verona, Italy
継続期間: 2018 9 52018 9 8

Other

Other6th International Conference on 3D Vision, 3DV 2018
CountryItaly
CityVerona
Period18/9/518/9/8

ASJC Scopus subject areas

  • Artificial Intelligence
  • Computer Science Applications
  • Computer Vision and Pattern Recognition

フィンガープリント 「Constant velocity 3D convolution」の研究トピックを掘り下げます。これらがまとまってユニークなフィンガープリントを構成します。

引用スタイル