A pytorch implementation of the "Selective Convolutional Descriptor Aggregation" algorithm

Go to file

CaoGang2018 4dcc749d98 men		2020-05-31 17:54:13 +08:00
.vscode	..	2020-05-31 10:27:56 +08:00
data	init	2020-05-29 18:05:22 +08:00
util	men	2020-05-31 17:54:13 +08:00
README.md	men	2020-05-31 17:54:13 +08:00
SCDA.py	cc for GMP	2020-05-31 10:30:19 +08:00
main.py	test code	2020-05-31 10:27:09 +08:00
requirements.txt	updata requirements	2020-05-29 19:32:34 +08:00

README.md

SCDA_pytorch

A pytorch implementation of the "Selective Convolutional Descriptor Aggregation" algorithm

NOTE

cpu-only version

no in [1]

train_data_L31a(i,:) = train_data_L31a(i,:) ./ norm(train_data_L31a(i,:));

largestConnectComponent

Details

install requirements

  pip install -r requirements.txt;

On CUB and split dataset in CUB_200.py.

random split CUB-200-2011 results:

	top1	top5
CUB	0.546	0.794

[1] Wei X S , Luo J H , Wu J , et al. Selective Convolutional Descriptor Aggregation for Fine-Grained Image Retrieval[J]. IEEE Transactions on Image Processing, 2017, 26(6):2868-2881.