Making a Case for Learning Motion Representations with Phase
Abstract
This work advocates Eulerian motion representation learning over the current standard Lagrangian optical flow model. Eulerian motion is well captured by using phase, as obtained by decomposing the image through a complex-steerable pyramid. We discuss the gain of Eulerian motion in a set of practical use cases: (i) action recognition, (ii) motion prediction in static images, (iii) motion transfer in static images and, (iv) motion transfer in video. For each task we motivate the phase-based direction and provide a possible approach.
Cite
Text
Pintea and van Gemert. "Making a Case for Learning Motion Representations with Phase." European Conference on Computer Vision, 2016. doi:10.1007/978-3-319-49409-8_8Markdown
[Pintea and van Gemert. "Making a Case for Learning Motion Representations with Phase." European Conference on Computer Vision, 2016.](https://mlanthology.org/eccv/2016/pintea2016eccv-making/) doi:10.1007/978-3-319-49409-8_8BibTeX
@inproceedings{pintea2016eccv-making,
title = {{Making a Case for Learning Motion Representations with Phase}},
author = {Pintea, Silvia L. and van Gemert, Jan C.},
booktitle = {European Conference on Computer Vision},
year = {2016},
pages = {55-64},
doi = {10.1007/978-3-319-49409-8_8},
url = {https://mlanthology.org/eccv/2016/pintea2016eccv-making/}
}