Anisotropic energy accumulation for stereoscopic image seam carving

Seam carving is an image resizing method that aims at adapting the image to various display screens while reducing the distortion as much as possible. Severe visual distortion may be introduced by repeated removal or insertion of seams within a concentrated region of the image. To reduce such visual distortion, we propose a new AESSC method. First, we distribute the energy of each pixel on the seam to its adjacent 8-connected pixels when removing or inserting a seam. Second, since the information of each pixel is anisotropic, we use the Sobel operator to detect the direction that has the maximum edge information and continue the energy accumulation along this direction. Besides, we incorporate a 3D structure consistency constraint in the energy function and adopt a pixel visibility maintenance method. Experimental results show that the proposed method can effectively reduce visual distortion for stereo images while maintaining the geometric consistency.

[1]  Takeshi Naemura,et al.  Seam carving for stereo images , 2010, 2010 3DTV-Conference: The True Vision - Capture, Transmission and Display of 3D Video.

[2]  Radomír Mech,et al.  Automatic Image Cropping using Visual Composition, Boundary Simplicity and Content Preservation Models , 2014, ACM Multimedia.

[3]  Pierre Vandergheynst,et al.  Stream carving: An adaptive seam carving algorithm , 2010, 2010 IEEE International Conference on Image Processing.

[4]  W. G. C. W. Kumara,et al.  Improved seam carving using meta-heuristics algorithms combination , 2015, 2015 Signal Processing and Intelligent Systems Conference (SPIS).

[5]  Bo Yan,et al.  Improved seam carving for image retargeting with sift feature preservation , 2015, 2015 IEEE China Summit and International Conference on Signal and Information Processing (ChinaSIP).

[6]  Matthew Turk,et al.  On Preserving Structure in Stereo Seam Carving , 2015, 2015 International Conference on 3D Vision.

[7]  Régis Vaillant,et al.  on Pattern Analysis and Machine Intelligence , 2005 .