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

Recent end-to-end deep neural networks for disparity regression have achievedthe state-of-the-art performance. However, many well-acknowledged specificproperties of disparity estimation are omitted in these deep learningalgorithms. Especially, matching cost volume, one of the most importantprocedure, is treated as a normal intermediate feature for the followingsoftargmin regression, lacking explicit constraints compared with thosetraditional algorithms. In this paper, inspired by previous canonicaldefinition of cost volume, we propose the noise-sampling cross entropy lossfunction to regularize the cost volume produced by deep neural networks to beunimodal and coherent. Extensive experiments validate that the proposednoise-sampling cross entropy loss can not only help neural networks learn moreinformative cost volume, but also lead to better stereo matching performancecompared with several representative algorithms.

Noise-Sampling Cross Entropy Loss: Improving Disparity Regression Via Cost Volume Aware Regularizer