PT Unknown
AU Victor Vaquero
   German Ros
   Francesc Moreno-Noguer
   Antonio Lopez
   Alberto Sanfeliu
TI Joint coarse-and-fine reasoning for deep optical flow
BT 24th International Conference on Image Processing
PY 2017
BP 2558
EP 2562
DI 10.1109/ICIP.2017.8296744
AB 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.
ER