北京生物医学工程

基于可伸缩窗视差生成的开放手术视频虚拟视点合成算法

Virtual view synthesis algorithm for operation video based on disparity generated with adaptive window

作者：贾博奇张梦诗梁楠武博张楠

单位：首都医科大学生物医学工程学院（北京 100069）

分类号：R318.04

出版年·卷·期（页码）：2018·37·6（603-610）

摘要：

目的本文为了提高开放手术视频的视差图质量, 减少虚拟视点中空洞, 提升虚拟视点主观质量, 提出基于可伸缩窗视差生成的虚拟视点合成算法, 为医学专家和医学生获得身临其境的手术示教和手术指导提供了可能。方法根据场景内容特点确定匹配窗口大小并进行立体匹配, 生成手术场景视差信息;基于深度绘制原理, 针对手术场景的特性填充空洞, 获得手术场景虚拟视点图像, 并对虚拟视点的主观质量进行评估。结果与固定窗方法相比较, 本文算法生成的虚拟视点图像中术野内物体边缘较光滑, 存在较少伪影。结论本文算法生成的手术虚拟视点视频取得了较高的主观质量, 为观看者提供了更好的观看体验。

Objective To improve the quality of disparity maps in open operation videos and to reduce holes in virtual views, the virtual view synthesis algorithm based on disparity generated with adaptive window is proposed. Methods According to the characteristics of the scene content, the size of the matching window is determined and stereo matching is performed to generate disparity information of the surgical scene. Based on depth image based render ( DIBR) , the holes are filled with the characteristics of the surgical scene, and then the subjective quality of the obtained virtual view is evaluated. Results Compared with the fix window method, the edges of the objects in the operative field in the virtual view generated by this algorithm are smooth and have fewer artifacts. Conclusions This algorithm can achieve a high subjective quality and offer a better viewing experience.

参考文献：

[1]吕源治, 王世刚, 任广旭, 等.二维多视点图像阵列采集及虚拟视点合成算法[J].哈尔滨工程大学学报, 2013, 34 (6) :763-767, 795.Lyu YZ, Wang SG, Ren GX, et al, Two dimensional multiview image array pickup and virtual viewpoint image synthesis algorithm[J]. Journal of Harbin Engineering University, 2013, 34 (6) :763-767, 795.

[2] Xu X, Po LM, Ng KH, et al. Depth map misalignment correction and dilation for DIBR view synthesis[J]. Signal Processing:Image Communication, 2013, 28 (9) :1023-1045.

[3] Sharma M, Chaudhury S, Lall B. A novel hybrid kinect-varietybased high-quality multiview rendering scheme for glass-free 3D displays[J]. IEEE Transactions on Circuits and Systems for Video Technology, 2017, 27 (10) :2098-2117.

[4]何贝, 王贵锦, 林行刚.结合Kinect深度图的快速视频抠图算法[J].清华大学学报 (自然科学版) , 2012, 52 (4) :561-565, 570.He B, Wang GJ, Lin HG. Quick matting for videos based on depth images of the Kinect[J]. Journal of Tsinghua University (Science and Technology) , 2012, 52 (4) :561-565, 570.

[5] Han CC, Hsiao HF. Depth estimation and video synthesis for 2D to 3D video conversion[J]. Journal of Signal Process Systems, 2014, 76:33-46.

[6] Gu Z, Su X, Liu Y, et al. Local stereo matching with adaptive support-weight, rank transform and disparity calibration[J].Pattern Recognition Letters, 2008, 29 (9) :1230-1235.

[7] Kanade T, Okutomi M. A stereo matching algorithm with an adaptive window:theory and experiment[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 1994, 16 (9) :920-932.

[8]张来刚, 魏仲慧, 何昕, 等.面向低纹理图像的快速立体匹配[J].液晶与显示, 2013, 28 (3) :450-458.Zhang LG, Wei ZH, He X, et al. New stereo matching method based edge extraction[J]. Chinese Journal of Liquid Crystals and Displays, 2013, 28 (3) :450-458.

[9]刘杰, 张建勋, 代煜, 等.基于跨尺度引导图像滤波的稠密立体匹配[J].光学学报, 2018, 38 (1) :0115004.Liu J, Zhang JX, Dai Y, et al. Dense Stereo Matching Based on Cross-Scale Guided Image Filtering[J]. Acta Optica Sinica, 2018, 38 (1) :0115004.

[10]汪辉, 陈芬, 焦任直, 等.基于空间加权的虚拟视点绘制算法[J].计算机工程与应用, 2016, 52 (8) :174-179.Wang H, Chen F, Jiao RZ, et al. Virtual view rendering algorithm based on spatial weighting[J]. Computer Engineering and Applications, 2016, 52 (8) :174-179.

[11] Criminisi A, Pérez P, Toyama K. Region filling and object removal by exemplar-based image inpainting[J]. IEEE Transactions on Image Processing, 2004, 13 (9) :1200-1212.

[12] Ahn I, Kim C. A novel depth-based virtual view synthesis method for free viewpoint video[J]. IEEE Transactions on Broadcasting, 2013, 59 (4) :614-626.

[13] Lai Y, Lan X, Liu Y, et al. An efficient depth image-based rendering with depth reliability maps for view synthesis[J].Journal of Visual Communication and Image Representation, 2016, 41:176-184.

[14] Schmeing M, Jiang X. Faithful Disocclusion Filling in Depth Image Based Rendering Using Superpixel-Based Inpainting[J].IEEE Transactions on Multimedia, 2015, 17 (12) :2160-2173.

[15]李超, 沈建新, 梁春.裸眼3D技术在显微手术中的研究与应用[J].计算机技术与发展, 2015, 25 (7) :170-173.Li C, Shen JX, Liang C. Research and application of naked-eye3D technology in microsurgery[J]. Computer Technology and Development, 2015, 25 (7) :170-173.

服务与反馈：

【文章下载】【加入收藏】

提示：您还未登录，请登录！点此登录