Abstract | ||
---|---|---|
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. |
Year | DOI | Venue |
---|---|---|
2018 | 10.1109/3DV.2018.00047 | 2018 International Conference on 3D Vision (3DV) |
Keywords | DocType | ISSN |
3D convolution,Constant velocity,Fourier transform | Journal | 2378-3826 |
ISBN | Citations | PageRank |
978-1-5386-8426-9 | 1 | 0.36 |
References | Authors | |
2 | 7 |
Name | Order | Citations | PageRank |
---|---|---|---|
Yusuke Sekikawa | 1 | 9 | 3.87 |
Kohta Ishikawa | 2 | 2 | 0.71 |
Hara, K. | 3 | 18 | 3.15 |
Yuichi Yoshida | 4 | 469 | 44.88 |
Koichiro Suzuki | 5 | 2 | 1.05 |
Ikuro Sato | 6 | 25 | 6.46 |
Hideo Saito | 7 | 1147 | 169.63 |