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

A great deal of research has demonstrated recently that multi-view stereo(MVS) matching can be solved with deep learning methods. However, these effortswere focused on close-range objects and only a very few of the deeplearning-based methods were specifically designed for large-scale 3D urbanreconstruction due to the lack of multi-view aerial image benchmarks. In thispaper, we present a synthetic aerial dataset, called the WHU dataset, wecreated for MVS tasks, which, to our knowledge, is the first large-scalemulti-view aerial dataset. It was generated from a highly accurate 3D digitalsurface model produced from thousands of real aerial images with precise cameraparameters. We also introduce in this paper a novel network, called RED-Net,for wide-range depth inference, which we developed from a recurrentencoder-decoder structure to regularize cost maps across depths and a 2D fullyconvolutional network as framework. RED-Net's low memory requirements and highperformance make it suitable for large-scale and highly accurate 3D Earthsurface reconstruction. Our experiments confirmed that not only did our methodexceed the current state-of-the-art MVS methods by more than 50% mean absoluteerror (MAE) with less memory and computational cost, but its efficiency aswell. It outperformed one of the best commercial software programs based onconventional methods, improving their efficiency 16 times over. Moreover, weproved that our RED-Net model pre-trained on the synthetic WHU dataset can beefficiently transferred to very different multi-view aerial image datasetswithout any fine-tuning. Dataset are available at http://gpcv.whu.edu.cn/data.

A Novel Recurrent Encoder-Decoder Structure for Large-Scale Multi-view Stereo Reconstruction from An Open Aerial Dataset