Sorry, you need to enable JavaScript to visit this website.

Domain Agnostic Video Prediction from Motion Selective Kernels

Citation Author(s):
Submitted by:
Veronique Prinet
Last updated:
19 September 2019 - 11:54pm
Document Type:
Presentation Slides
Document Year:
2019
Event:
Presenters Name:
Da Li
Paper Code:
3335
Categories:

Abstract 

Abstract: 

Existing conditional video prediction approaches train a network from large databases and generalise to previously unseen data. We take the opposite stance, and introduce a model that learns from the first frames of a given video and extends its content and motion, to, \eg double its length. To this end, we propose a dual network that can use in a flexible way both dynamic and static convolutional motion kernels, to predict future frames. We demonstrate experimentally the robustness of our approach on challenging videos in-the-wild and show that it is competitive related baselines.

up
0 users have voted:

Dataset Files

prinet-icip2019-presentation-onlineversion.pdf

(138)