基于空间和运动图像视频显著性检测.pdfVIP

Image and Video Region Saliency Based on Space and Motion Jian Li, Martin Levine, Xiangjing An, Zhenping Sun, and Hangen He Abstract. This paper proposes a new bottom-up paradigm for detecting visual saliency in images and videos, which is based on scale space analysis of the log amplitude spectrum of natural images and videos. A salient region is assumed to be any region exhibiting a distinct pattern whose intensity, color, texture and motion is different from the rest of the image. Thus patterns which appear frequently as well as uniform regions are suppressed to produce salient region pop-out. We show that the convolution of the image log amplitude spectrum with a low-pass Gaussian ker- nel (at the proper scale) is equivalent to such an image saliency detector. A saliency map can then be obtained by reconstructing the 2-D signal using the original phase spectrum and an appropriately filtered log amplitude spectrum to produce pop-out. The optimal scale for each image feature channel (intensity, color, motion) is de- termined by minimizing the entropy of its saliency map. This produces four maps which are then fused by a weighted linear combination. Significantly, the approach does not require the setting of any parameters. We demonstrate experimentally that the proposed model has the ability to highlight small and large salient regions and to inhibit repeating patterns in both images and videos. 1 Introduction Object detection in images and videos has been and continues to be of great interest in computer vision. One aspect regarding such computational methods deals with the issue of whether to sample images densely or at certain salient points. The latter is not only perceptually plausible, but also would reduce computing costs [1, 2, 3, 4, 5]. Researchers have always realized that locating salient regions in a scene is a Jian Li · Xiangjing An · Zhenping Sun · Hangen He Institute of Automation, Natl. Univ. of Defense Technol