9-1-图像分割-概述幻灯片.ppt

* * * * 图像分割 图像识别与人工智能研究所,多谱信息处理国家重点实验室 陶文兵 华中科技大学图像识别与人工智能研究所 多谱信息处理技术国家重点实验室 分割的目的和意义 图像分割是计算机视觉研究中的基础问题和难点之一 图像分割就是把图像分成各具特性的区域并提取出感兴趣目标 图像分割的难点和挑战性 对一般图像中的大量视觉模型进行建模的复杂性 图像理解本身的内在模糊性 当没有一个明确的任务来指导注意机制 * 图像工程的三层模型 image segmentation … Goal: Break up the image into meaningful or perceptually similar regions Types of segmentations Oversegmentation Undersegmentation Multiple Segmentations Segmentation as a result Rother et al. 2004 Segmentation for efficiency [Felzenszwalb and Huttenlocher 2004] [Hoiem et al. 2005, Mori 2005] [Shi and Malik 2001] Segments as primitives for recognition J. Tighe and S. Lazebnik, submitted to ECCV 2010 Major processes for segmentation Bottom-up: group tokens with similar features Top-down: group tokens that likely belong to the same object [Levin and Weiss 2006] Bottom-up segmentation Group together similar-looking pixels for efficiency of further processing “Bottom-up” process Unsupervised X. Ren and J. Malik. Learning a classification model for segmentation. ICCV 2003. “superpixels” The goals of segmentation Separate image into coherent “objects” “Bottom-up” or “top-down” process? Supervised or unsupervised? Berkeley segmentation database:/Research/Projects/CS/vision/grouping/segbench/ image human segmentation Top-down segmentation E. Borenstein and S. Ullman, “Class-specific, top-down segmentation,” ECCV 2002 A. Levin and Y. Weiss, “Learning to Combine Bottom-Up and Top-Down Segmentation,” ECCV 2006. Top-down segmentation E. Borenstein and S. Ullman, “Class-specific, top-down segmentation,” ECCV 2002 A. Levin and Y. Weiss, “Learning to Combine Bottom-Up and Top-Down Segmentation,” ECCV 2006. Normalized cuts Top-down segmentation 图像分割方法的发展现状 目前图像分割方法主要有三个比较重要的方向： 基于统计理论的图像分割方法 Mean Shift，DDMCMC… 基于变分模型的图像分割方法 Snake Model，GAC，M-S Model，C-V Model… 基于图论的图像分割方法 Graph Cuts，Normalize Cuts… * Three basic theory in Image Segmentation Statistics Variational Graph Two basic Model in Image Segmentation Statistics