Paper Title
Image Retargeting Quality Assessment: A Backward Registration Approach

The paper shows the result of large scale study of image retargeting quality on a collection of images obtained by explanatory image retargeting methods. Apart from many Image Quality Assessment (IQA) metrics, the quality degradation during image retargeting occurs due to artificial modifications and the difficulty for Image Retargeting Quality Assessment lies in the variation of image resolution and content, which makes it difficult to straightaway evaluate the class of degradation like traditional IQA. The paper explains image retargeting by integrating the structure of resampling, grid generation and forward sampling. Also, that the geometric change estimation is an effective way to simplify the connection amongst the original image and the retargeted image. We frame the geometric change estimation as the Backward Registration problem and extract the fast features and freak descriptors and assign a productive solution Keywords: Backward Registration, Fast Feature, Geometric change, Image Fusion, Image retargeting quality assessment