Effective retargeting for image coding

Tingxiao Hu, Bo Yan

Abstract: An effective retargeting scheme is presented in this paper in order to improve the efficiency of image coding. Firstly, we propose to calculate the information loss in image retargeting based on learning. Then before encoding an image, we divide the image into several grids and retarget each grid according to our learning results. The retargeted image can be encoded by the traditional image codec. In the receiver side, with the help of the side information, the compressed image can be decoded and reconstructed. Simulation results show that the proposed scheme is able to provide much higher compression performance compared with the JPEG method.