期刊文献+

Kinect深度图像快速修复算法 被引量:22

Fast Inpainting Algorithm for Kinect Depth Map
下载PDF
导出
摘要 深度提取是基于"纹理+深度"自由立体视频系统的关键技术,而立体视频实际应用系统需要高效快速的深度图提取.提出一种针对Kinect提取深度图的快速修复算法.首先,对Kinect提取的彩色纹理图和深度图进行对齐裁剪,并采用背景填充算法对裁剪后的深度图进行初步修复;然后,对初步修复后的深度图进行基于颜色匹配的快速修复,得到质量较好的可用深度图.实验结果表明,本算法能有效修复原始深度图中由于遮挡而引起的空洞,获取的深度图整体平滑度好、边缘清晰;在普通PC机上达到25~30帧/s的处理帧率,实现了深度图的实时提取. Depth extraction is a key step in 3D video system based on texture plus depth. Real application systems require high efficiency and fast depth extraction. This paper presents a fast inpainting algorithm for Kinect depth map. This paper clips and aligns the origin color and depth images captured by Kinect, and partially fill holes in the clipped depth image using a background based method. Then this paper uses a fast inpainting algorithm based on color match to fill the remaining holes in the depth image to obtain a better depth image. Experimental results show that the proposed algorithm can efficiently repair errors in the original depth map such as holes caused by occlusion. Finally we obtain a smooth depth map with clear edges. The processing speed can reach 25 -30 frame/s by using an ordinary desktop computer to realize real-time depth map extraction.
出处 《上海大学学报(自然科学版)》 CAS CSCD 北大核心 2012年第5期454-458,共5页 Journal of Shanghai University:Natural Science Edition
基金 国家自然科学基金资助项目(60832003 61172096 61171084) 上海市科委重点资助项目(10510500500) 上海市教委重点资助项目(09ZZ90)
关键词 深度图修复 实时获取 颜色匹配 KINECT 立体视频 depth map inpainting real-time capture color match Kinect stereo video
  • 相关文献

参考文献9

  • 1DARIBO I, SAITO H. A novel inpainting-based layered depth video for 3DTV [ J ]. IEEE Transactions on Broadcasting, 2011:533-541.
  • 2UM G M, KIM T, BANG G. Multi-view 3D video acquisition using hybrid camera with beem spliter [ C ]// 3 DTV Conference : The True Vision-Capture, Transmission and Display of 3D Video (3DTV-CON). 2011:1-4.
  • 3DARIBO I, TILLIER C, PESQUET-POPESCU B. Distance dependent depth filtering in 3D warping for 3DTV [ C]// Multimedia Signal Processing. 2007:312-315.
  • 4CHEN W Y, CHANG Y L, LIN S F, et al. Efficient depth image based rendering with edge dependent depth filter and interpolation [ C ] // IEEE International Conference on Multimedia and Expo. 2005:1314-1317.
  • 5KIM S Y, CHO J H, KOSCHAN A, et al. Spatial and temporal enhancement of depth images captured by a time-of-flight depth sensor [ C ] // Pattern Recognition (ICPR). 2010:2358-2361.
  • 6ZHU J J, WANG L, YANG R G, et al. Fusion of time-of- flight depth and stereo for high accuracy depth maps [C]//Computer Vision and Pattern Recognition ( CVPR2008 ). 2008 : 1-8.
  • 7WILSON A. Using a depth camera as a touch sensor [C]// ITS' 10 ACM International Conference on Interactive Tabletops and Surfaces. 2010:69-72.
  • 8MATYUNIN S, VATOLIN D, BERDNIKOV Y. Temporal filtering for depth maps generated by kinect depth camera [ C ]//The True Vision-Capture, Transmission and Display of 3 D Video (3 DTV-CON ). 2001 : 1-4.
  • 9Jl X P, WEI Z Q, FENG Y W. Effective vehicle detection technique for traffic surveillance systems [ J ]. Visual Communication & Image Representation, 2005, 17(3) :647-658.

同被引文献163

  • 1李彬,刘冀伟,韩鸿哲,李正熙.复杂背景下人体骨架的提取[J].微计算机信息,2004,20(7):43-44. 被引量:11
  • 2丁益洪,平西建,胡敏.基于随机Hough变换的深度图像分割[J].计算机辅助设计与图形学学报,2005,17(5):902-907. 被引量:13
  • 3李久权,王平,王永强.CT图像分割几种算法[J].微计算机信息,2006,22(02S):240-242. 被引量:29
  • 4马登武,叶文,李瑛.基于包围盒的碰撞检测算法综述[J].系统仿真学报,2006,18(4):1058-1061. 被引量:111
  • 5赵清杰,钱芳,蔡利栋译.计算机视觉[M].北京:机械工业出版社.2005:311-322.
  • 6狄海进.基于三维视觉的手势跟踪及人机交互中的应用[D].南京大学,2011.
  • 7Fu Jingjing, Wang Shiqi, Lu Yan, et al. Kinect-like depth denoising[C]//Cireuits and Systems ISCAS). Seoul: IEEE, 2012:512-515.
  • 8Massimo Camplani, Luis Salgado. Efficient spatio-tempo- ral hole filling strategy for kineet depth maps[C]//Three- Dimensional Image Processing (3DIP) and Applications II. California: SPIE, 2012:82-90.
  • 9Milani S, Calvagno G. Joint denoising and interpola- tion of depth maps for MS Kinect sensors[C]//Acous- ties, Speech and Signal Processing (ICASSP). Koyo- to: IEEE, 2012:797-800.
  • 10Xu Kang, Zhou Jun, Wang Zhen. A method of hole- filling for the depth map generated by kinect with mov- ing objects detection[C]//Broadband Multimedia Sys- tems and Broadcasting (BMSB). Seoul: IEEE, 2012.

引证文献22

二级引证文献106

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部