基于卷积神经网络的视差图生成技术

doi:10.11772/j.issn.1001-9081.2017071659

计算机应用 ›› 2018, Vol. 38 ›› Issue (1): 255-259.DOI: 10.11772/j.issn.1001-9081.2017071659

• 虚拟现实与多媒体计算 • 上一篇下一篇

基于卷积神经网络的视差图生成技术

朱俊鹏¹, 赵洪利², 杨海涛³

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

收稿日期:2017-07-05 修回日期:2017-09-03 出版日期:2018-01-10 发布日期:2018-01-22
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:
装备学院校级基础研究项目（DXZT-JC-ZZ-2013-009）。

Disparity map generation technology based on convolutional neural network

ZHU Junpeng¹, ZHAO Hongli², YANG Haitao³

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).

摘要/Abstract

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

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)

中图分类号:

TP391.413

朱俊鹏, 赵洪利, 杨海涛. 基于卷积神经网络的视差图生成技术[J]. 计算机应用, 2018, 38(1): 255-259.

ZHU Junpeng, ZHAO Hongli, YANG Haitao. Disparity map generation technology based on convolutional neural network[J]. Journal of Computer Applications, 2018, 38(1): 255-259.

参考文献

[1] 刘建伟,刘媛,罗雄麟.深度学习研究进展[J].计算机应用研究,2014,31(7):1921-1930.(LIU J W, LIU Y, LUO X L. Research and development on deep learning[J]. Application Research of Computers, 2014, 31(7):1921-1930.)
[2] 赵天奇.裸眼3D内容生成和显示若干关键技术研究[D].北京:北京邮电大学,2015:22-43.(ZHAO T Q. Research on key technologies of naked eye there-dimensional display and its content generation[D]. Beijing:Beijing University of Posts and Telecommunications, 2015:22-43.)
[3] 李博乐.基于DIBR的裸眼3D显示系统研究与实现[D].重庆:重庆大学,2015:7-40.(LI B L. Research and implementation of glasses-free 3D display system based on DIBR[D]. Chongqing:Chongqing University, 2015:7-40.)
[4] 谭伟敏.裸眼3D显示关键技术研究[D].重庆:重庆大学,2014:32-43.(TAN W M. Research on key technologies of glasses-free 3D display[D]. Chongqing:Chongqing University, 2014:32-42.)
[5] 李彦冬,郝宗波,雷航.卷积神经网络研究综述[J].计算机应用,2016,36(9):2508-2515.(LI Y D, HAO Z B, LEI H. Survey of convolutional neural network[J]. Journal of Computer Applications, 2016, 36(9):2508-2515.)
[6] 卢宏涛,张秦川.深度卷积神经网络在计算机视觉中的应用研究综述[J].数据采集与处理,2016,31(1):1-17(LU H T, ZHANG Q C. Application of deep convolutional neural network in computer vision[J]. Journal of Data Acquisition and Processing, 2016, 31(1):1-17)
[7] SILVER D, HUANG A, MADDISON C J, et al. Mastering the game of go with deep neural networks and tree search[J]. Nature, 2016, 529(7587):484-489.
[8] ZEILER M D, FERGUS R. Stochastic pooling for regularization of deep convolutional neural networks[EB/OL].[2017-01-11]. http://www.matthrwzeiler.com/pubs/iclr2013/iclr2013.pdf.
[9] MURPHY K P. Machine Learning:A Probabilistic Perspective[M]. Cambridge, MA:MIT Press, 2012:82-92.
[10] TATARCHENKO M, DOSOVITSKIY A, BROX T. Single-view to multi-view:reconstructing unseen views with a convolutional network[J]. Knowledge & Information Systems, 2015, 38(1):231-257.
[11] DOSOVITSKIY A, FISCHER P, ILG E, et al. FlowNet:learning optical flow with convolutional networks[C]//Proceedings of the 2015 IEEE International Conference on Computer Vision. Washington, DC:IEEE Computer Society, 2015:2758-2766.
[12] RICHTER S R, VINEET V, ROTH S, et al. Playing for data:ground truth from computer games[C]//Proceedings of the 2016 European Conference on Computer Vision. Berlin:Springer, 2016:102-118.
[13] GEIGER A, LENZ P, STILLER C, et al. Vision meets robotics:the KITTI dataset[J]. International Journal of Robotics Research, 2013, 32(11):1231-1237.
[14] WANG C, YAN X, SMITH M, et al. A unified framework for automatic wound segmentation and analysis with deep convolutional neural networks[C]//EMBC 2015:Proceedings of the 37th Annual International Conference of the IEEE Engineering in Medicine and Biology Society. Piscataway, NJ:IEEE, 2015:2415-2418.
[15] HE K, ZHANG X, REN S, et al. Delving deep into rectifiers:surpassing human-level performance on ImageNet classification[C]//Proceedings of the 2016 IEEE International Conference on Computer Vision. Piscataway, NJ:IEEE, 2016:1026-1034.
[16] ATHEY S, IMBENS G. Machine learning methods for estimating heterogeneous causal effects[J]. Statistics, 2015, 113(27):7353-7360.

基于卷积神经网络的视差图生成技术

Disparity map generation technology based on convolutional neural network

PDF

可视化

摘要/Abstract

引用本文

使用本文

参考文献

相关文章 15

编辑推荐

Metrics

[1]	王贺兵, 张春梅. 基于非对称卷积-压缩激发-次代残差网络的人脸关键点检测[J]. 计算机应用, 2021, 41(9): 2741-2747.
[2]	郑志强, 胡鑫, 翁智, 王雨禾, 程曦. 基于改进DenseNet的牛眼图像特征提取方法[J]. 计算机应用, 2021, 41(9): 2780-2784.
[3]	宋中山, 梁家锐, 郑禄, 刘振宇, 帖军. 基于双向门控尺度特征融合的遥感场景分类[J]. 计算机应用, 2021, 41(9): 2726-2735.
[4]	李康康, 张静. 基于注意力机制的多层次编码和解码的图像描述模型[J]. 计算机应用, 2021, 41(9): 2504-2509.
[5]	张永斌, 常文欣, 孙连山, 张航. 基于字典的域名生成算法生成域名的检测方法[J]. 计算机应用, 2021, 41(9): 2609-2614.
[6]	赵宏, 孔东一. 图像特征注意力与自适应注意力融合的图像内容中文描述[J]. 计算机应用, 2021, 41(9): 2496-2503.
[7]	徐江浪, 李林燕, 万新军, 胡伏原. 结合目标检测的室内场景识别方法[J]. 计算机应用, 2021, 41(9): 2720-2725.
[8]	牟长宁, 王海鹏, 周丕宇, 侯鑫行. 基于图卷积神经网络的串联质谱从头测序[J]. 计算机应用, 2021, 41(9): 2773-2779.
[9]	曾祥银, 郑伯川, 刘丹. 基于深度卷积神经网络和聚类的左右轨道线检测[J]. 计算机应用, 2021, 41(8): 2324-2329.
[10]	曹玉红, 徐海, 刘荪傲, 王紫霄, 李宏亮. 基于深度学习的医学影像分割研究综述[J]. 计算机应用, 2021, 41(8): 2273-2287.
[11]	秦斌斌, 彭良康, 卢向明, 钱江波. 司机分心驾驶检测研究进展[J]. 计算机应用, 2021, 41(8): 2330-2337.
[12]	黄程程, 董霄霄, 李钊. 基于二维Winograd算法的深流水线5×5卷积方法[J]. 计算机应用, 2021, 41(8): 2258-2264.
[13]	吴则举, 焦翠娟, 陈亮. 基于改进Faster R-CNN的轮胎缺陷检测方法[J]. 计算机应用, 2021, 41(7): 1939-1946.
[14]	杨粟, 欧阳智, 杜逆索. 基于相关度距离的无监督并行哈希图像检索[J]. 计算机应用, 2021, 41(7): 1902-1907.
[15]	武光利, 李雷霆, 郭振洲, 王成祥. 基于改进的双向长短期记忆网络的视频摘要生成模型[J]. 计算机应用, 2021, 41(7): 1908-1914.