| 摘要 |
A method and apparatus are provided for encoding/decoding an image based on a region of interest (ROI). The method and apparatus provide error-resilience by duplicating ROI data in a variable manner according to image features or by reducing non-ROI data. The ROI based image encoding apparatus and method estimate a complexity for each block located in the ROI of an image, transform the image by creating more duplicate blocks for ROI blocks having high complexity and fewer duplicate blocks for ROI blocks having low complexity, and encode the transformed image.
|
