Journal of Computer Applications ›› 2018, Vol. 38 ›› Issue (1): 255-259.DOI: 10.11772/j.issn.1001-9081.2017071659

Previous Articles     Next Articles

Disparity map generation technology based on convolutional neural network

ZHU Junpeng1, ZHAO Hongli2, YANG Haitao3   

  1. 1. Department of Graduate Management, Equipment Academy, Beijing 101416, China;
    2. Training Department, Equipment Academy, Beijing 101416, China;
    3. Complex Electronic System Simulation Laboratory, Equipment Academy, Beijing 101416, China
  • Received:2017-07-05 Revised:2017-09-03 Online:2018-01-10 Published:2018-01-22
  • Supported by:
    This work is partially supported by the Academy of Equipment School Level Basic Research Project (DXZT-JC-ZZ-2013-009).


朱俊鹏1, 赵洪利2, 杨海涛3   

  1. 1. 装备学院 研究生管理大队, 北京 101416;
    2. 装备学院 训练部, 北京 101416;
    3. 装备学院 复杂电子系统仿真实验室, 北京 101416
  • 通讯作者: 朱俊鹏
  • 作者简介:朱俊鹏(1993-),男(土家族),湖南张家界人,硕士研究生,主要研究方向:信息网络安全;赵洪利(1964-),男,北京人,教授,博士,主要研究方向:信息网络安全;杨海涛(1979-),男,山东烟台人,副研究员,博士,主要研究方向:信息网络安全。
  • 基金资助:

Abstract: Focusing on the issues such as high cost, long time consumption and background holes in the disparity map in naked-eye 3D applications, learning and prediction algorithm based on Convolutional Neural Network (CNN) was introduced. Firstly, the change rules of a dataset could be mastered through training and learning the dataset. Secondly, the depth map with continuous lasting depth value was attained by extracting and predicting the features of the left view in the input CNN. Finally, the right view was produced by the superposition of diverse stereo pairs after folding the predicted depth and original maps. The simulation results show that the pixel-wise reconstruction error of the proposed algorithm is 12.82% and 10.52% lower than that of 3D horizontal disparity algorithm and depth image-based rendering algorithm. In addition, the problems of background hole and background adhesion have been greatly improved. The experimental results show that CNN can improve the image quality of disparity maps.

Key words: naked-eye 3D, disparity map, background hole, feature extraction, Convolutional Neural Network (CNN)

摘要: 针对裸眼三维中视差图生成过程中存在的高成本、长耗时以及容易出现背景空洞的问题,提出了一种基于卷积神经网络(CNN)学习预测的算法。首先通过对数据集的训练学习,掌握数据集中的变化规律;然后对输入卷积神经网络中的左视图进行特征提取和预测,得到深度值连续的深度图像;其次将预测所得到的每一个深度图和原图进行卷积,将生成的多个立体图像对进行叠加,最终形成右视图。仿真结果表明:该算法的像素重构尺寸误差相比基于水平视差的三维显示算法和深度图像视点绘制的算法降低了12.82%和10.52%,且背景空洞、背景粘连等问题都得到了明显改善。实验结果表明,卷积神经网络能提高视差图生成的图像质量。

关键词: 裸眼三维, 视差图, 背景空洞, 特征提取, 卷积神经网络

CLC Number: