Bilinear Attention Networks for Person Retrieval

Pengfei Fang, Jieming Zhou, Soumava Kumar Roy, Lars Petersson, Mehrtash Harandi; Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV), 2019, pp. 8030-8039

Abstract


This paper investigates a novel Bilinear attention (Bi-attention) block, which discovers and uses second order statistical information in an input feature map, for the purpose of person retrieval. The Bi-attention block uses bilinear pooling to model the local pairwise feature interactions along each channel, while preserving the spatial structural information. We propose an Attention in Attention (AiA) mechanism to build inter-dependency among the second order local and global features with the intent to make better use of, or pay more attention to, such higher order statistical relationships. The proposed network, equipped with the proposed Bi-attention is referred to as Bilinear ATtention network (BAT-net). Our approach outperforms current state-of-the-art by a considerable margin across the standard benchmark datasets (e.g., CUHK03, Market-1501, DukeMTMC-reID and MSMT17).

Related Material


[pdf] [supp]
[bibtex]
@InProceedings{Fang_2019_ICCV,
author = {Fang, Pengfei and Zhou, Jieming and Roy, Soumava Kumar and Petersson, Lars and Harandi, Mehrtash},
title = {Bilinear Attention Networks for Person Retrieval},
booktitle = {Proceedings of the IEEE/CVF International Conference on Computer Vision (ICCV)},
month = {October},
year = {2019}
}