VIDEO SALIENCY DETECTION IN FREQUENCY DOMAIN USING MODIFIED HFT
Keywords:
Hypercomplex Algorithm, Saliency DetectionAbstract
As it is known that the ability of sensory system of human being to seek out salient object in image which is quite fast. However, machine modeling of this basic intelligent
behavior still remains a challenge. Image as well as video saliency detection technique based on neural network is presented in this paper to distinct salient regions with their surroundings. Even though several existing saliency detection algorithms have been proposed whereas the obtained saliency
maps have not given the satisfying results. In this paper the problem of visual saliency is tackled. First of all in this technique saliency detection is considered as a frequency domain analysis. Secondly, algorithm accomplished this by using the concept of non-saliency. Third, proposed algorithm tends to at an equivalent time consider the detection of salient regions of varied size. In this paper a replacement of bottom-up approach for detection of visual saliency, characterized by a scale-space analysis of the amplitude spectrum of natural pictures and videos. A modified Hypercomplex Fourier is proposed and experimental analysis demonstrates that model gives efficient results and achieved the accuracy of about 80%.
References
R. Achanta, F. Estrada, P. Wils, and S. Susstrunk, “Salient regiondetection and segmentation,” In International Conf. On ComputerVision Systems, 2008.
R. Achanta, S. Hemami, F. Estrada, and S. Susstrunk, “Frequencytunedsalient region detection,” In IEEE CVPR, p. 1597C1604, 2009.
A. Borji, “Boosting bottom-up and top-down visual features forsaliency estimation,” In IEEE CVPR, 2012.
A. Borji, D. N. Sihite, and L. Itti, “Salient object detection: Abenchmark,” In Proc. European Conference on Computer Vision(ECCV), 2012.
Ming-Ming Cheng, Jonathan Warrell, Wen-Yan Lin, Shuai Zheng, andVibhav Vineet, “Efficient salient region detection with soft imageabstraction,” In ICCV, 2013.
Bowen Jiang, Lihe Zhang, Huchuan Lu, and Minghsuan Yang,“Saliency detection via absorbing markov chain,” In ICCV, 2013.
Chuan Yang, Lihe Zhang, Huchuan Lu, Xiang Ruan, and Ming-HsuanYang, “Saliencydetection via graph-based manifold ranking,” In IEEECVPR, 2013.
Z. Zivkovic, “Improved adaptive gaussian mixture model forbackground subtraction,” In ICPR, 2004.
C. K. J. Harel and P. Perona, “Graph-based visual saliency”,Proceedings of Neural Information Processing Systems (NIPS), 2006.
Yuanyuan Dong, Mahsa T. Pourazad, and Panos Nasiopoulos, “HumanVisual System-based Saliency Detection For HDR Content, " IEEETransactions On Multimedia, Vol. 18, No. 4, April 2016.
Wenliang Qiu Xinbo Gao Bing Han, “A superpixel-based CRF saliencydetection approach”, Elsevier, 2017.
B. Schauerte and R. Stiefelhagen, “Predicting human gaze usingquaternion dct image signature saliency and face detection”, InProceedings of the 12th IEEE Workshop on the Applications ofComputer Vision (WACV) / IEEE Winter Vision Meetings,Breckenridge, pp. pp 9–11, January 2012.
