Constant velocity 3D convolution

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

Research output: Chapter in Book/Report/Conference proceedingConference contribution

4 Citations (Scopus)


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.

Original languageEnglish
Title of host publicationProceedings - 2018 International Conference on 3D Vision, 3DV 2018
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages9
ISBN (Electronic)9781538684252
Publication statusPublished - 2018 Oct 12
Event6th International Conference on 3D Vision, 3DV 2018 - Verona, Italy
Duration: 2018 Sept 52018 Sept 8


Other6th International Conference on 3D Vision, 3DV 2018


  • 3D convolution
  • Constant velocity
  • Fourier transform

ASJC Scopus subject areas

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


Dive into the research topics of 'Constant velocity 3D convolution'. Together they form a unique fingerprint.

Cite this