Reproduction of paper:Learning Discriminative Features with Multiple Granularities for Person Re-Identification
DarkNet after modifying the source code.
(1)In the caffe2darknet conversion, need to modify the DarkNet source code to add the slice layer.
(2)The source code needs to be modified to allow the max_pool layer to support multi-scale kernel size and stride size
(3)Modify DarkNet source code to add support for numpy format for picture input in Python interface.
@ARTICLE{2018arXiv180401438W,
author = {{Wang}, G. and {Yuan}, Y. and {Chen}, X. and {Li}, J. and {Zhou}, X.},
title = "{Learning Discriminative Features with Multiple Granularities for Person Re-Identification}",
journal = {ArXiv e-prints},
archivePrefix = "arXiv",
eprint = {1804.01438},
primaryClass = "cs.CV",
keywords = {Computer Science - Computer Vision and Pattern Recognition},
year = 2018,
month = apr,
adsurl = {http://adsabs.harvard.edu/abs/2018arXiv180401438W},
adsnote = {Provided by the SAO/NASA Astrophysics Data System}
}