Open zdx3578 opened 5 years ago
Hi, thanks for your interest! We have not thought of this yet. As PFF generates per-pixel kernels warping input for the output, instead of producing the output directly, the input and output are expected to be in the same domain. But we do expect some future direction such as using PFF to produce optical flows with further constraints.
thanks!
can we use Predictive Filter Flow for depth predcition task ?
ref paper: Adversarial Collaboration: Joint Unsupervised Learning of Depth, Camera Motion, Optical Flow and Motion Segmentation https://arxiv.org/abs/1805.09806
depth related motion; motion related flow