English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

We investigate two crucial and closely related aspects of CNNs for opticalflow estimation: models and training. First, we design a compact but effectiveCNN model, called PWC-Net, according to simple and well-established principles:pyramidal processing, warping, and cost volume processing. PWC-Net is 17 timessmaller in size, 2 times faster in inference, and 11\% more accurate on Sintelfinal than the recent FlowNet2 model. It is the winning entry in the opticalflow competition of the robust vision challenge. Next, we experimentallyanalyze the sources of our performance gains. In particular, we use the sametraining procedure of PWC-Net to retrain FlowNetC, a sub-network of FlowNet2.The retrained FlowNetC is 56\% more accurate on Sintel final than thepreviously trained one and even 5\% more accurate than the FlowNet2 model. Wefurther improve the training procedure and increase the accuracy of PWC-Net onSintel by 10\% and on KITTI 2012 and 2015 by 20\%. Our newly trained modelparameters and training protocols will be available onhttps://github.com/NVlabs/PWC-Net

Models Matter, So Does Training: An Empirical Study of CNNs for Optical Flow Estimation