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

Both optical flow and stereo disparities are image matches and can thereforebenefit from joint training. Depth and 3D motion provide geometric rather thanphotometric information and can further improve optical flow. Accordingly, wedesign a first network that estimates flow and disparity jointly and is trainedwithout supervision. A second network, trained with optical flow from the firstas pseudo-labels, takes disparities from the first network, estimates 3D rigidmotion at every pixel, and reconstructs optical flow again. A final stage fusesthe outputs from the two networks. In contrast with previous methods that onlyconsider camera motion, our method also estimates the rigid motions of dynamicobjects, which are of key interest in applications. This leads to betteroptical flow with visibly more detailed occlusions and object boundaries as aresult. Our unsupervised pipeline achieves 7.36% optical flow error on theKITTI-2015 benchmark and outperforms the previous state-of-the-art 9.38% by awide margin. It also achieves slightly better or comparable stereo depthresults. Code will be made available.

UFD-PRiME: Unsupervised Joint Learning of Optical Flow and Stereo Depth through Pixel-Level Rigid Motion Estimation