Paper 1: Beyond Bags of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories
文献信息:
概念界定:SPM Spatial Pyramid Matching. The resulting "spatial pyramid" is a simple and conputationally efficient extension of an orderless bag-of-features image representation.
research gap: bag-of-features methods, which represent an image as an orderless collection of local features, have recently demonstrated impressive levels of performance. However, because these methods disregard all information about the spatial layout of the features, they have severely limited descriptive ability. In particular, they are incapable of capturing shape or of segmenting an object from its background.
目标问题:两个集合的比较问题,这两个集合所含的特征数不同,而且特征之间无序。
创新点:This paper presents a method for recognizing scene categories based on approximate global geometric correspondence.
研究方法:This trchnique works by partitioning the image into increasingly fine sub-regions and computing histograms of local features found inside each subregion.
研究目的:
作者观点:
引用例证:
研究意义:
Github:
其他:
特征提取:通常使用dense-SIFT,就是在图片上撒一个网格,每个网格中心点都计算一个SIFT(128维)的表达,将图片表示成一堆SIFT的集合。
字典训练:K-Means
.
标签:features,Notes,Scene,SIFT,they,local,image,Recognition From: https://www.cnblogs.com/zhaoke271828/p/17235815.html