Classical Local Descriptors

Classical local descriptors refer to those were proposed many years ago but have a profound influence on the development of local image description as well as related applications. Scale-Invariant Feature Transform (SIFT) and Speeded Up Robust Feature (SURF) are the two widely used descriptors in computer vision. Especially for SIFT, it is an extremely popular solution to various applications, ranging from object recognition, image retrieval, to structure from motion, etc. While for SURF, it is a first and predominant choice for those applications requiring fast or near real-time image matching until the very recent flourish of binary descriptors. Another classical local feature is Local Binary Pattern (LBP) proposed in the 1990s. Along with many variants, LBP has been ubiquitous in texture classification and many face-related tasks, e.g., face recognition, face detection, and facial expression recognition. Because of their popularity, we choose to introduce them in detail in this chapter.

[1]  G LoweDavid,et al.  Distinctive Image Features from Scale-Invariant Keypoints , 2004 .

[2]  Bin Fan,et al.  Local Intensity Order Pattern for feature description , 2011, 2011 International Conference on Computer Vision.

[3]  J. Friedman.  Special Invited Paper-Additive logistic regression: A statistical view of boosting , 2000 .

[4]  Matti Pietikäinen,et al.  Modeling pixel process with scale invariant local patterns for background subtraction in complex scenes , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[5]  Shengcai Liao,et al.  Learning Multi-scale Block Local Binary Patterns for Face Recognition , 2007, ICB.

[6]  Wen Gao,et al.  Local Gabor binary pattern histogram sequence (LGBPHS): a novel non-statistical model for face representation and recognition , 2005, Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1.

[7]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).