VCNet: A generative model for volume completion
VCNet: A generative model for volume completion
复制标题
DOI：
10.1016/j.visinf.2022.04.004
复制
发表时间：
2022-04
期刊：
影响因子：
--
通讯作者：
Jun Han;Chaoli Wang
中科院分区：
文献类型：
--
作者：
Jun Han;Chaoli Wang
我们提出了VCNet，一种新的深度学习方法，通过合成缺失的子卷来完成卷。我们的解决方案利用了生成对抗网络（GAN），该网络可以学习使用对抗和体积损失来完成体积。VCNet的核心设计具有扩张的残留块和长期连接。在训练期间，VCNet首先随机掩蔽基本子体积（例如，长方体，切片）从完整的卷，并学会恢复它们。此外，我们设计了一个两阶段的算法来稳定和加速网络优化。一旦经过训练，VCNet将不完整的卷作为输入，并自动识别和高质量地填充缺失的子卷。我们定量和定性测试VCNet与体积数据集的各种特性，以证明其有效性。我们还将VCNet与基于扩散的解决方案和两种基于GAN的解决方案进行了比较。
We present VCNet, a new deep learning approach for volume completion by synthesizing missing subvolumes. Our solution leverages a generative adversarial network (GAN) that learns to complete volumes using the adversarial and volumetric losses. The core design of VCNet features a dilated residual block and long-term connection. During training, VCNet first randomly masks basic subvolumes (e.g., cuboids, slices) from complete volumes and learns to recover them. Moreover, we design a two-stage algorithm for stabilizing and accelerating network optimization. Once trained, VCNet takes an incomplete volume as input and automatically identifies and fills in the missing subvolumes with high quality. We quantitatively and qualitatively test VCNet with volumetric data sets of various characteristics to demonstrate its effectiveness. We also compare VCNet against a diffusion-based solution and two GAN-based solutions.