Distinctive Image Features from Scale-Invariant Keypoints
International Journal of Computer Vision
Seam carving for content-aware image resizing
ACM SIGGRAPH 2007 papers
A Linear Time Histogram Metric for Improved SIFT Matching
ECCV '08 Proceedings of the 10th European Conference on Computer Vision: Part III
A comparative study of image retargeting
ACM SIGGRAPH Asia 2010 papers
Using eye-tracking to assess different image retargeting methods
Proceedings of the ACM SIGGRAPH Symposium on Applied Perception in Graphics and Visualization
Image quality assessment: from error visibility to structural similarity
IEEE Transactions on Image Processing
| Hi-index | 0.00 |
The prosperity of image retargeting technique leads to the significant need of effective approaches for assessing different image retargeting methods. In this paper, we propose a novel automatic approach providing human perception based image retargeting assessment by measuring salient region similarity between the original image and the target image. First, the salient regions in the original image and the target image are matched using gradient, color and direction features. Then, the similarity between the salient regions are measured based on feature distances and region saliency. Based on salient region similarity, the quality of target image is assessed with two criteria derived from real user requirements in retargeting, important content retainment and visual artifact reduction, and the overall scores of the target image quality are finally calculated by integrating assessment results on different criteria. Experiments demonstrate the effectiveness of the proposed approach.