虽然计算机视觉已经全部转向Deep Learning，但还是想写几篇博文对过去十年里一直使用的Bag of Visual Word模型进行归纳总结。
针对BoW模型，网上也能搜索到大量资料，下面的几个文章是其中比较具有代表性的。
2004--ECCV-"Visual Categorization with Bags of Keypoints"
2010--CVPR--"Locality-constrained Linear Coding for image classification"
2010--CVPR--"Aggregating local descriptors into a compact image representation"
2011--BMVC--"The devil is in the details: an evaluation of recent feature encoding methods"
2013--IJCV--"Image Classification with the Fisher Vector: Theory and Practice"
2014--TPAMI--"Feature Coding in Image Classification: A Comprehensive Study"
2014--BMVC--Tutorial--"Image Representations, from shallow to deep"
1. Basic BoW
2. Spatial Pyramid Model
3. Fisher Vector
4. Sparse Coding
原文：http://www.cnblogs.com/ZhimingLuo/p/4941336.html