多尺度级联网络的墓室壁画数字生成技术

Digital Generation Technology for Tomb Murals Based on Multiscale Cascade Network

因墓道狭窄,分块揭取的大型墓室壁画在高清采集后会存在周边信息缺失的情况.为了能够重建壁画分块间丢失的数据,提出了一种多尺度级联网络的墓室壁画数字生成技术.首先,对墓室壁画进行大尺度生成;然后,将重建结果输入深度语义小尺度生成网络中进行精细的数字信息生成.在小尺度生成网络引入自注意力机制,增强生成区域与全局信息的关联性,解决生成区域边界的伪影问题.在反馈损失中改进纹理损失,提高重建信息纹理精细度以及壁画生成效果.为加快训练进程、促进梯度反向传播效率,在生成网络中加入跳跃连接.通过消融实验和对比组实验验证,该数字生成技术可提高壁画分块外延信息的纹理匹配率并弱化伪影的影响,在客观指标峰值信噪比和结构相似度上均取得较好的结果.

Largescale tomb murals are divided into several blocks by narrow passages.Hence,during highdefinition collection,certain information around these blocks may be missing.To address this,a digital generation technology for tomb murals based on a multiscale cascade network is proposed,for reconstructing these lost data between mural blocks.In this approach,tomb murals were first generated on a large scale,and the reconstruction results were subsequently input into a deep semantic smallscale generation network to generate fine digital information.A selfattention mechanism was introduced into the smallscale generation network to enhance the correlation between the generation region and the global information and solve the artifact problem of the generation region boundaries.In terms of the feedback loss,the texture loss and the texture fineness of the reconstructed information are improved,and the mural generation effect is also improved.Jump connections were added to the generation network to accelerate the training process and enhance the efficiency of gradient backpropagation.Based on ablation and comparative group experiments,the proposed digital generation technology can improve the texture matching rate of mural block epitaxial information and reduce the influence of artifacts.This proposed method achieves good objective indexes for the peak signaltonoise ratio and structural similarity.