转载自博客：

深度学习在CV领域的进展以及一些由深度学习演变的新技术

https://blog.****.net/wzz18191171661/article/details/70161595 

特征检测与匹配（块）

经典模型：

TILDE: A Temporally Invariant Learned DEtector 
https://arxiv.org/pdf/1411.4568.pdf

MatchNet: Unifying Feature and Metric Learning for Patch-Based Matching 
https://pdfs.semanticscholar.org/81b9/24da33b9500a2477532fd53f01df00113972.pdf

Discriminative Learning of Deep Convolutional Feature Point Descriptors 
http://cvlabwww.epfl.ch/~trulls/pdf/iccv-2015-deepdesc.pdf

Learning to Assign Orientations to Feature Points 
https://arxiv.org/pdf/1511.04273.pdf

PN-Net: Conjoined Triple Deep Network for Learning Local Image Descriptors 
https://arxiv.org/pdf/1601.05030.pdf

Multi-scale Pyramid Pooling for Deep Convolutional Representation 
http://ieeexplore.ieee.org/stamp/stamp.jsp?arnumber=7301274

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition 
https://arxiv.org/pdf/1406.4729.pdf

Learning to Compare Image Patches via Convolutional Neural Networks 
https://arxiv.org/pdf/1504.03641.pdf

PixelNet: Representation of the pixels, by the pixels, and for the pixels 
http://www.cs.cmu.edu/~aayushb/pixelNet/pixelnet.pdf

LIFT: Learned Invariant Feature Transform 
https://arxiv.org/pdf/1603.09114.pdf

TILDE: A Temporally Invariant Learned DEtector： 

MatchNet: Unifying Feature and Metric Learning for Patch-Based Matching： 

Discriminative Learning of Deep Convolutional Feature Point Descriptors： 

Learning to Assign Orientations to Feature Points： 

PN-Net: Conjoined Triple Deep Network for Learning Local Image Descriptors： 

Multi-scale Pyramid Pooling for Deep Convolutional Representation： 

Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition： 

Learning to Compare Image Patches via Convolutional Neural Networks： 

PixelNet: Representation of the pixels, by the pixels, and for the pixels：

 

LIFT: Learned Invariant Feature Transform：