A Novel Multiresolution Spatiotemporal Saliency Detection Model and Its Applications in Image and Video Compression

• Published in: IEEE transactions on image processing Vol. 19; no. 1; pp. 185 - 198
• Main Authors: Guo, Chenlei, Zhang, Liming
• Format: Journal Article
• Language: English
• Published: New York, NY IEEE 01.01.2010; Institute of Electrical and Electronics Engineers
• Subjects: Applied sciences; Artificial intelligence; Coding, codes; Computational efficiency; Computational modeling; Computer science; control theory; systems; Computer vision; Exact sciences and technology; Fourier transforms; Hierarchical selectivity (HS); Image coding; Image processing; Image resolution; Information, signal and communications theory; multiresolution wavelet domain foveation (MWDF); Pattern recognition; Pattern recognition. Digital image processing. Computational geometry; phase spectrum of Fourier transform (PFT); phase spectrum of quaternion Fourier transform (PQFT); Quaternions; receiver operating characteristic (ROC) curve; Signal and communications theory; Signal processing; Spatiotemporal phenomena; spatiotemporal saliency map; Strontium; Telecommunications and information theory; Video compression; visual attention; Video coding; Performance evaluation; State of the art; Natural scene; Fourier transformation; Foveation; Image processing; Quaternion; Data compression; multiresolution wavelet domain foveation (MWDF); Selectivity; phase spectrum of Fourier transform (PFT); Multiresolution analysis; Video signal processing; phase spectrum of quaternion Fourier transform (PQFT); Image coding; spatiotemporal saliency map; receiver operating characteristic (ROC) curve; Localization; Cost lowering; Computer vision; Image resolution; Amplitude spectrum; Image compression; Tree structure; Computational complexity; Image quality; Hierarchical selectivity (HS); Object detection; Visual attention

Salient areas in natural scenes are generally regarded as areas which the human eye will typically focus on, and finding these areas is the key step in object detection. In computer vision, many models have been proposed to simulate the behavior of eyes such as SaliencyToolBox (STB), Neuromorphic Vi...