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

Convolutional neural networks(CNN) have been shown to perform better than theconventional stereo algorithms for stereo estimation. Numerous efforts focus onthe pixel-wise matching cost computation, which is the important building blockfor many start-of-the-art algorithms. However, those architectures are limitedto small and single scale receptive fields and use traditional methods for costaggregation or even ignore cost aggregation. Differently we take them both intoconsideration. Firstly, we propose a new multi-scale matching cost computationsub-network, in which two different sizes of receptive fields are implementedparallelly. In this way, the network can make the best use of both variants andbalance the trade-off between the increase of receptive field and the loss ofdetail. Furthermore, we show that our multi-dimension aggregation sub-networkwhich containing 2D convolution and 3D convolution operations can provide richcontext and semantic information for estimating an accurate initial disparity.Finally, experiments on challenging stereo benchmark KITTI demonstrate that theproposed method can achieve competitive results even without any additionalpost-processing.

Cascaded multi-scale and multi-dimension convolutional neural network for stereo matching