Efficient Light Field Image Coding with Depth Estimation and View Synthesis

Efficient light field image coding method is proposed based on conversion to multi-view image, depth estimation, multi-view coding, and view synthesis. Firstly, compatibility of light field image and multi-view image is discussed, and then depth estimation method based on texture-edge-aware horizontal and vertical view matching and depth-smoothing is explained. Secondly, a view-synthesis method from up to four reference views is proposed, which adopts depth-base occlusion hole inpainting. Finally, by combining these methods together with a hierarchical bi-directional inter-view coding of multi-view image and depth maps, coding results are reported.

[1]  Peter Lambert,et al.  Steered mixture-of-experts for light field coding, depth estimation, and processing , 2017, 2017 IEEE International Conference on Multimedia and Expo (ICME).

[2]  Cristian Perra,et al.  High efficiency coding of light field images based on tiling and pseudo-temporal data arrangement , 2016, 2016 IEEE International Conference on Multimedia & Expo Workshops (ICMEW).

[3]  Takanori Senoh,et al.  Fast depth estimation using non-iterative local optimization for super multi-view images , 2015, 2015 IEEE Global Conference on Signal and Information Processing (GlobalSIP).

[4]  E. Adelson,et al.  The Plenoptic Function and the Elements of Early Vision , 1991 .

[5]  Changick Kim,et al.  Depth-Based Disocclusion Filling for Virtual View Synthesis , 2012, 2012 IEEE International Conference on Multimedia and Expo.