1 paper
Jiaxing Wang, Jihua Zhu, Shanmin Pang +3
Aggregating deep convolutional features into a global image vector has attracted sustained attention in image retrieval. In this paper, we propose an efficient unsupervised aggrega…