中国邮电高校学报(英文版) ›› 2022, Vol. 29 ›› Issue (5): 83-91.doi: 10.19682/j.cnki.1005-8885.2022.0005

• Artificial Intelligence • 上一篇    下一篇

TSR: algorithm of image hole-filling based on three-step repairing

李富澄1,邓军勇2,朱筠1,罗佳莹1,任含1,3   

  1. 1. 西安邮电大学
    2. 西安邮电学院
    3.
  • 收稿日期:2021-05-31 修回日期:2021-09-16 出版日期:2022-10-31 发布日期:2022-10-28
  • 通讯作者: 李富澄 E-mail:15619234469@163.com
  • 基金资助:
    面上项目 效能驱动的光互连视频阵列处理器动态自重构体系结构; 三维光电混合片上网络关键技术研究;性能驱动可编程自重构图形处理器体系结构研究;数据驱动轻核阵列处理器自重构机制研究;可编程动态自重构三维阵列芯片体系结构关键技术;陕西省科技统筹;陕西省重点研发计划

TSR: algorithm of image hole-filling based on three-step repairing

Li Fucheng, Deng Junyong, Zhu Yun, Luo Jiaying, Ren Han   

  • Received:2021-05-31 Revised:2021-09-16 Online:2022-10-31 Published:2022-10-28

摘要:

In order to solve the hole-filling mismatch problem in virtual view synthesis, a three-step repairing (TSR) algorithm was proposed. Firstly, the image with marked holes is decomposed by the non-subsampled shear wave transform ( NSST), which will generate high-/ low-frequency sub-images with different resolutions. Then the improved Criminisi algorithm was used to repair the texture information in the high-frequency sub-images, while the improved curvature driven diffusion (CDD) algorithm was used to repair the low-frequency sub-images with the image structure information. Finally, the repaired parts of high-frequency and low-frequency sub-images are synthesized to obtain the final image through inverse NSST. Experiments show that the peak signal-to-noise ratio (PSNR) of the TSR algorithm is improved by an average of 2 - 3 dB and 1 - 2 dB compared with the Criminisi algorithm and the nearest neighbor interpolation (NNI) algorithm, respectively.

关键词: virtual view point synthesis| hole-filling| three-step repairing (TSR)| Criminisi algorithm| curvature driven diffusions (CDD) algorithm

Abstract:

In order to solve the hole-filling mismatch problem in virtual view synthesis, a three-step repairing (TSR) algorithm was proposed. Firstly, the image with marked holes is decomposed by the non-subsampled shear wave transform ( NSST), which will generate high-/ low-frequency sub-images with different resolutions. Then the improved Criminisi algorithm was used to repair the texture information in the high-frequency sub-images, while the improved curvature driven diffusion (CDD) algorithm was used to repair the low-frequency sub-images with the image structure information. Finally, the repaired parts of high-frequency and low-frequency sub-images are synthesized to obtain the final image through inverse NSST. Experiments show that the peak signal-to-noise ratio (PSNR) of the TSR algorithm is improved by an average of 2 - 3 dB and 1 - 2 dB compared with the Criminisi algorithm and the nearest neighbor interpolation (NNI) algorithm, respectively.

Key words: virtual view point synthesis| hole-filling| three-step repairing (TSR)| Criminisi algorithm| curvature driven diffusions (CDD) algorithm