Supplementary Information-Large Scale High-Resolution Land Cover Mapping with Multi-Resolution Data

Supplementary Information-Large Scale High-Resolution Land Cover Mapping with Multi-Resolution Data
复制标题

DOI:
--
复制
发表时间:
2019
期刊:
--
影响因子:
--
通讯作者:
Caleb Robinson;Le Hou;Nikolay Malkin;Rachel Soobitsky;Jacob Czawlytko;B. Dilkina;N. Jojic
Caleb Robinson;Le Hou;Nikolay Malkin;Rachel Soobitsky;Jacob Czawlytko;B. Dilkina;N. Jojic
中科院分区:
其他
文献类型:
--
作者:
Caleb Robinson;Le Hou;Nikolay Malkin;Rachel Soobitsky;Jacob Czawlytko;B. Dilkina;N. Jojic

文献摘要

被引文献

相似文献

在这里,我们提供了在主论文中描述的Web应用程序的描述和屏幕截图。用户可以使用预先计算的模型输出或在服务器上计算的动态模型输出在现有底图(在本例中为ESRI World Imagery或OpenStreetMap)上进行“绘制”。用户可以更改绘制的土地覆盖层的不透明度,以查看它如何与观察到的图像对齐,更改类权重以创建更好的拟合,并提交更正以包含在未来几轮模型训练中或由其他用户加载。参见https://aka.ms/cvprlandcover。
Here we provide a description and screenshots of the web application described in the main paper. Users are able to “paint” over existing basemaps (in this case the ESRI World Imagery or OpenStreetMap) with either precomputed model output or on-the-fly model output computed on a server. Users can change the opacity of the painted land cover layer to see how it aligns with the observed imagery, change class weights to create a better fit, and submit the corrections to be included in future rounds of model training or loaded by other users. See https://aka.ms/cvprlandcover.