Abstract

This document introduces the background and the usage of the DunhuangGrottoes Dataset and the benchmark. The documentation first starts with thebackground of the Dunhuang Grotto, which is widely recognised as an pricelessheritage. Given that digital method is the modern trend for heritage protectionand restoration. Follow the trend, we release the first public dataset forDunhuang Grotto Painting restoration. The rest of the documentation details thepainting data generation. To enable a data driven fashion, this datasetprovided a large number of training and testing example which is sufficient fora deep learning approach. The detailed usage of the dataset as well as thebenchmark is described.

 

Quick Read (beta)