Fast depth image denoising and enhancement using a deep convolutional network
Fast depth image denoising and enhancement using a deep convolutional network
复制标题
DOI：
10.1109/icassp.2016.7472127
复制
发表时间：
2016-03
期刊：
影响因子：
--
通讯作者：
Xin Zhang;Ruiyuan Wu
中科院分区：
文献类型：
--
作者：
Xin Zhang;Ruiyuan Wu
提出了一种基于轻卷积网络的深度图像去噪和增强框架。该网络包含三层，分别用于高维投影、缺失数据补全和图像重建。我们联合使用深度和视觉图像作为输入。对于灰度图像，我们设计了一种预处理程序来增强边缘，去除不必要的细节。对于深度图像，我们提出了一种数据增强策略来再生和增加基本训练数据。此外，我们提出了一种加权损失函数用于网络训练，以自适应提高学习效率。我们在基准数据上测试了我们的算法，并在实时速度下获得了非常有希望的视觉和定量结果。
We propose a depth image denoising and enhancement framework using a light convolutional network. The network contains three layers for high dimension projection, missing data completion and image reconstruction. We jointly use both depth and visual images as inputs. For the gray image, we design a pre-processing procedure to enhance the edges and remove unnecessary detail. For the depth image, we propose a data augmentation strategy to regenerate and increase essential training data. Further, we propose a weighted loss function for network training to adaptively improve the learning efficiency. We tested our algorithm on benchmark data and obtained very promising visual and quantitative results at real-time speed.