In this paper an effective and efficient method for contentawareimage resizing is proposed. It is based on the solutionof a linear system where each pixel displacement (compressionor expansion) is determined in dependence of the visualrelevance of the pixel itself.The linear nature of the model allows real-time application ofthe method even for large images.This fully automatic approach can be also improved by interactivelyproviding cues such as geometric constraints and/ormanual relevant object labeling.The results have proven that the presented method achievesresults comparable or superior to existent strategies, while improvingefficiency.
|Number of pages||4|
|Publication status||Published - 2010|
All Science Journal Classification (ASJC) codes
- Computer Vision and Pattern Recognition
- Signal Processing