基于稀疏贝叶斯学习的遮挡场景深度恢复方法
淦相琦 曾蝶 吴婷

重庆对外经贸学院，重庆，401520；

摘要：本文提出一种基于稀疏贝叶斯学习的光场深度恢复方法，针对遮挡场景中传统方法存在的边界模糊和伪影问题，通过建立层次化概率模型将遮挡问题转化为稀疏信号恢复任务。该方法利用自适应稀疏约束学习和贝叶斯推断机制，自动识别并滤除遮挡噪声，同时保持有效深度信息，在复杂遮挡场景下实现了深度估计误差降低，展现出优异的边缘保持能力和鲁棒性，为光场相机的三维感知及虚拟现实、自动驾驶等应用提供了新的技术解决方案。

关键词：稀疏贝叶斯学习；遮挡场景；深度恢复；光场成像

参考文献

[1]MaiYiWei,“Reducing plenoptic camera artifacts,”Computer Graphics Forum,vol.29,no.6,pp.1955–1968, 2010.

[2]T.E.Bishop and P.Favaro,“The light field camera:Extended depth of field,aliasing, and superresolution.”IEEE Trans.Pattern Anal.Mach.Intell.(PAMI),vol.34,no.5,pp.972–986,2012.

[3]C.Perwass and P.Wietzke,“Single lens 3d-camera with extended depthof-field,”in SPIE Elect.Imaging,2012.

[4]H.Lin,C.Chen,S.B.Kang,and J.Yu,“Depth recovery from light field using focal stack symmetry.” in Proceedings of the IEEE International Conference on Computer Vision(ICCV),2015.

[5]M. W.Tao,S.Hadap,J.Malik,and R.Ramamoorthi,“Depth from combining defocus and correspondence using light-field cameras[C],”in Proc.IEEE Int.Conf.Comput.Vis.,2013,pp.673–680.

[6]Yang D G,Xiao Z L,Yang H,et al.Depth estimation from light field analysis based multiple cues fusion[J]. Chinese Journal of Computers, 2015,38(12):2437-2449.

[7]S.Wanner and B.Goldluecke,“Globally consistent depth labelingof 4D light fields[C],”in Proc.IEEE Conf.Comput.Vis.Pattern Recog.,2012,pp.41–48.

[8]H.Lin,C.Chen,S.B.Kang,and J.Yu,“Depth recovery from light field using focal stack symmetry.”in Proceedings of the IEEE International Conference on Computer Vision (ICCV),2015.