最优化网格形变的图像非均匀映射方法
|
摘要
本文针对内容感知的图像非均匀映射问题,建立了一种有效的基于网格形变的全局最优化模型。引入了一系列保护重要网格形变的能量约束项,使得当图像尺寸比例发生改变时,视觉显著的重要区域不发生形变,并且尽可能地保留图像中的全局信息,给予用户最佳的图像浏览体验。此外,也通过建立网格初始划分问题的最优化数学模型,实现覆盖重要区域的网格密集分布,减少关键对象边界信息的扭曲失真。通过与其他主流方法的结果比较与分析,证明本文方法能够更好的保证图像关键区域不发生形变,得到质量更优的结果图像。
We built an effective global optimization model of image retargeting based on grid deformation. This model introduces a series of energy functions that prevent important grid deforming. The proposed approach allows resizing images into arbitrary aspect ratio while protecting visually prominent area and retaining the global information as much as possible,in order to give users the best experience of browsing image. In addition,we achieved that goal that important grid distribute densely by establishing the optimization model of partitioning the initial grid in the original image,reducing the artifacts on the boundary of the prominent object. A number of comparison experiments show that our method performs better than stateof-the-art approaches.
We built an effective global optimization model of image retargeting based on grid deformation. This model introduces a series of energy functions that prevent important grid deforming. The proposed approach allows resizing images into arbitrary aspect ratio while protecting visually prominent area and retaining the global information as much as possible,in order to give users the best experience of browsing image. In addition,we achieved that goal that important grid distribute densely by establishing the optimization model of partitioning the initial grid in the original image,reducing the artifacts on the boundary of the prominent object. A number of comparison experiments show that our method performs better than stateof-the-art approaches.
引文
[1]Avidan S,Shamir A.Seam carving for content-aware image resizing[J].ACM Transactions on Graphics,2007,6(3):10-18.
[2]Rubinstein M,Shamir A,Avidan S.Improved seam carving for video retargeting[J].ACM Transactions on Graphics,2008,27(3):1-9.
[3]Grundmann M,Kwatra V,Han M,et al.Discontinuous seam-carving for video retargeting[C]∥2010 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).San Francisco,2010:569-576.
[4]Wu L,Cao L,Chen C W.Fast and improved seam carving with strip partition and neighboring probability constraints[C]∥2013 IEEE International Symposium on Circuits and Systems(ISCAS).2013:2812-2815.
[5]Botao Wang,Hongkai Xiong,Zhiquan Ren,Chang Wen Chen.Deformable Shape Preserving Video Retargeting With Salient Curve Matching[J].IEEE Journal on Emerging and Selected Topics in Circuits and Systems,2014,4(1):82-94.
[6]Wolf L,Guttmann M,Cohen-Or D.Non-homogeneous content-driven video-retargeting[C]∥In Proceedings of the Eleventh IEEE International Conference on Computer Vision,Rio de Janeiro,2007:1-6.
[7]Wang Y S,Tai C L,Sorkine O,et al.Optimized scaleand-stretch for image resizing[C]∥ACM Transactions on Graphics(TOG),2008,27(5):118.
[8]Chen R,Freedman D,Karni Z,et al.Content-aware image resizing by quadratic programming[C]∥IEEE Computer Vision and Pattern Recognition Workshops(CVPRW),2010:1-8.
[9]Hu W,Luo Z,Fan X.Image Retargeting Via Adaptive Scaling With Geometry Preservation[J].IEEE Journal on Emerging and Selected Topics in Circuits and Systems,2014,4(1):70-81.
[10]Goferman S,Zelnik-Manor L,Tal A.Context-aware saliency detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(10):1915-1926.
[11]ALGLIB Project.Bound and linear equality/inequality constrained optimization[EB/OL].http:∥www.alglib.net/optimization/boundandlinearlyconstrained.php,2014.
[12]Vaquero D,Turk M,Pulli K,et al.A survey of image retargeting techniques[C]∥In Proceedings of SPIE,2010,7798:779814.
[2]Rubinstein M,Shamir A,Avidan S.Improved seam carving for video retargeting[J].ACM Transactions on Graphics,2008,27(3):1-9.
[3]Grundmann M,Kwatra V,Han M,et al.Discontinuous seam-carving for video retargeting[C]∥2010 IEEE Conference on Computer Vision and Pattern Recognition(CVPR).San Francisco,2010:569-576.
[4]Wu L,Cao L,Chen C W.Fast and improved seam carving with strip partition and neighboring probability constraints[C]∥2013 IEEE International Symposium on Circuits and Systems(ISCAS).2013:2812-2815.
[5]Botao Wang,Hongkai Xiong,Zhiquan Ren,Chang Wen Chen.Deformable Shape Preserving Video Retargeting With Salient Curve Matching[J].IEEE Journal on Emerging and Selected Topics in Circuits and Systems,2014,4(1):82-94.
[6]Wolf L,Guttmann M,Cohen-Or D.Non-homogeneous content-driven video-retargeting[C]∥In Proceedings of the Eleventh IEEE International Conference on Computer Vision,Rio de Janeiro,2007:1-6.
[7]Wang Y S,Tai C L,Sorkine O,et al.Optimized scaleand-stretch for image resizing[C]∥ACM Transactions on Graphics(TOG),2008,27(5):118.
[8]Chen R,Freedman D,Karni Z,et al.Content-aware image resizing by quadratic programming[C]∥IEEE Computer Vision and Pattern Recognition Workshops(CVPRW),2010:1-8.
[9]Hu W,Luo Z,Fan X.Image Retargeting Via Adaptive Scaling With Geometry Preservation[J].IEEE Journal on Emerging and Selected Topics in Circuits and Systems,2014,4(1):70-81.
[10]Goferman S,Zelnik-Manor L,Tal A.Context-aware saliency detection[J].IEEE Transactions on Pattern Analysis and Machine Intelligence,2012,34(10):1915-1926.
[11]ALGLIB Project.Bound and linear equality/inequality constrained optimization[EB/OL].http:∥www.alglib.net/optimization/boundandlinearlyconstrained.php,2014.
[12]Vaquero D,Turk M,Pulli K,et al.A survey of image retargeting techniques[C]∥In Proceedings of SPIE,2010,7798:779814.