Publication
Joint coarse-and-fine reasoning for deep optical flow
Conference Article
Conference
IEEE International Conference on Image Processing (ICIP)
Edition
24th
Pages
2558-2562
Doc link
http://dx.doi.org/10.1109/ICIP.2017.8296744
File
Authors
-
Vaquero Gomez, Victor
-
Ros, German
-
Moreno Noguer, Francesc
-
M. Lopez, Antonio
-
Sanfeliu Cortés, Alberto
Projects associated
Abstract
We propose a novel representation for dense pixel-wise estimation tasks using CNNs that boosts accuracy and reduces training time, by explicitly exploiting joint coarse-and-fine reasoning. The coarse reasoning is performed over a discrete classification space to obtain a general rough solution, while the fine details of the solution are obtained over a continuous regression space. In our approach both components are jointly estimated, which proved to be beneficial for improving estimation accuracy. Additionally, we propose a new network architecture, which combines coarse and fine components by treating the fine estimation as a refinement built on top of the coarse solution, and therefore adding details to the general prediction. We apply our approach to the challenging problem of optical flow estimation and empirically validate it against state-of-the-art CNN-based solutions trained from scratch and tested on large optical flow datasets.
Categories
computer vision, feature extraction, pattern classification.
Author keywords
optical flow, convolutional neural net- works, regression, classification
Scientific reference
V. Vaquero, G. Ros, F. Moreno-Noguer, A. M. and A. Sanfeliu. Joint coarse-and-fine reasoning for deep optical flow, 24th IEEE International Conference on Image Processing, 2017, Beijing, pp. 2558-2562.
Follow us!