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

Due to the domain differences and unbalanced disparity distribution acrossmultiple datasets, current stereo matching approaches are commonly limited to aspecific dataset and generalize poorly to others. Such domain shift issue isusually addressed by substantial adaptation on costly target-domainground-truth data, which cannot be easily obtained in practical settings. Inthis paper, we propose to dig into uncertainty estimation for robust stereomatching. Specifically, to balance the disparity distribution, we employ apixel-level uncertainty estimation to adaptively adjust the next stagedisparity searching space, in this way driving the network progressively pruneout the space of unlikely correspondences. Then, to solve the limited groundtruth data, an uncertainty-based pseudo-label is proposed to adapt thepre-trained model to the new domain, where pixel-level and area-leveluncertainty estimation are proposed to filter out the high-uncertainty pixelsof predicted disparity maps and generate sparse while reliable pseudo-labels toalign the domain gap. Experimentally, our method shows strong cross-domain,adapt, and joint generalization and obtains \textbf{1st} place on the stereotask of Robust Vision Challenge 2020. Additionally, our uncertainty-basedpseudo-labels can be extended to train monocular depth estimation networks inan unsupervised way and even achieves comparable performance with thesupervised methods. The code will be available athttps://github.com/gallenszl/UCFNet.

Digging Into Uncertainty-based Pseudo-label for Robust Stereo Matching