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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Large-scale synthetic datasets are beneficial to stereo matching but usuallyintroduce known domain bias. Although unsupervised image-to-image translationnetworks represented by CycleGAN show great potential in dealing with domaingap, it is non-trivial to generalize this method to stereo matching due to theproblem of pixel distortion and stereo mismatch after translation. In thispaper, we propose an end-to-end training framework with domain translation andstereo matching networks to tackle this challenge. First, joint optimizationbetween domain translation and stereo matching networks in our end-to-endframework makes the former facilitate the latter one to the maximum extent.Second, this framework introduces two novel losses, i.e., bidirectionalmulti-scale feature re-projection loss and correlation consistency loss, tohelp translate all synthetic stereo images into realistic ones as well asmaintain epipolar constraints. The effective combination of above twocontributions leads to impressive stereo-consistent translation and disparityestimation accuracy. In addition, a mode seeking regularization term is addedto endow the synthetic-to-real translation results with higher fine-graineddiversity. Extensive experiments demonstrate the effectiveness of the proposedframework on bridging the synthetic-to-real domain gap on stereo matching.

StereoGAN: Bridging Synthetic-to-Real Domain Gap by Joint Optimization of Domain Translation and Stereo Matching