Fusing Generic Objectness and Visual Saliency for Salient Object DetectionInternational Conference on Computer Vision (ICCV) 2011
Source codeThis source code contains:
P. Felzenszwalb and D. Huttenlocher. Efficient Graph-Based Image Segmentation, IJCV 2004. I wrap the function for Matlab. B. Alexe, T. Deselaers and V. Ferrari. What is an object?, CVPR 2010. S. Goferman, L. Zelnik Manor and A. Tal. Context-Aware Saliency Detection, CVPR 2010. K.-Y. Chang, T.-L. Liu, H.-T. Chen and S.-H. Lai. Fusing Generic Objectness and Visual Saliency for Salient Object Detection, ICCV 2011. T. Judd, K. Ehinger, F. Durand, and A. Torralba. Learning to predict where humans look, ICCV 2009. The mex files are built under the VC6 (32-bit) and the Visual Studio 2010 (64-bit) environments. You may need to rebuild them if you use any other version of Visual Studio or other OS.
To get a quick overview: RequirementMSRA datasetWe treat the scores ([1], [1]\Saliency, Ours-Obj, and Ours-Sal-Obj) as the weights of windows and use the expected windows to evaluate the precision, recall and F-measure according to the description in [2].
PASCAL VOC 2007We apply non maximum suppression (intersection-over-union>0.7) to each setting ([1], [1]\Saliency, Ours-Obj, and Ours-Sal-Obj) first and then evaluate the mAP value according to VOC criteria.
Download
Bibtex@InProceedings{Chang11, author = {Kai-Yueh Chang and Tyng-Luh Liu and Hwann-Tzong Chen and Shang-Hong Lai}, title = {Fusing Generic Objectness and Visual Saliency for Salient Object Detection}, booktitle = {IEEE International Conference on Computer Vision (ICCV)}, year = {2011} } Reference
[1] B. Alexe, T. Deselaers and V. Ferrari. What is an object? CVPR 2010. | ||||||||||||||||||||||||||||||