TY - JOUR
T1 - Towards multi-scale deep features learning with correlation metric for person re-identification
AU - Zhu, Dandan
AU - Zhou, Qiangqiang
AU - Han, Tian
AU - Chen, Yongqing
AU - Zhao, Defang
AU - Yang, Xiaokang
N1 - Publisher Copyright:
© 2020 Elsevier B.V.
PY - 2021/2/15
Y1 - 2021/2/15
N2 - Previous person re-identification (Re-ID) methods usually focus on extracted features to against the appearance variations of pedestrians under different circumstances. The problem caused by the scale variations did not attract much attention and is not well addressed either. In this work, we propose a novel Multi-scale Deep Feature Learning with correlation metric (MDFLCM) model to handle the scale problem in Re-ID. Specifically, multi-scale high-level features are extracted by a specially designed end-to-end multi-scale deep convolutional network (MS-DCN) at various resolution levels. By adding an extra correlation layer in our MDFLCM model, we can achieve the accuracy of image patch matching up to pixel-wise level. Different from other methods extracting multi-scale features through multiple networks, we extract multi-scale features via a single network with one input image. Extensive comparative evaluations with state-of-the-art methods on four public datasets: CUHK01, CUHK03, Market 1501 and DukeMTMC-reID, demonstrate the effectiveness of the proposed MDFLCM model on Re-ID.
AB - Previous person re-identification (Re-ID) methods usually focus on extracted features to against the appearance variations of pedestrians under different circumstances. The problem caused by the scale variations did not attract much attention and is not well addressed either. In this work, we propose a novel Multi-scale Deep Feature Learning with correlation metric (MDFLCM) model to handle the scale problem in Re-ID. Specifically, multi-scale high-level features are extracted by a specially designed end-to-end multi-scale deep convolutional network (MS-DCN) at various resolution levels. By adding an extra correlation layer in our MDFLCM model, we can achieve the accuracy of image patch matching up to pixel-wise level. Different from other methods extracting multi-scale features through multiple networks, we extract multi-scale features via a single network with one input image. Extensive comparative evaluations with state-of-the-art methods on four public datasets: CUHK01, CUHK03, Market 1501 and DukeMTMC-reID, demonstrate the effectiveness of the proposed MDFLCM model on Re-ID.
KW - Correlation layer
KW - Deep convolutional network
KW - Multi-scale features
KW - Pixel-wise level
UR - http://www.scopus.com/inward/record.url?scp=85098699247&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85098699247&partnerID=8YFLogxK
U2 - 10.1016/j.knosys.2020.106675
DO - 10.1016/j.knosys.2020.106675
M3 - Article
AN - SCOPUS:85098699247
SN - 0950-7051
VL - 213
JO - Knowledge-Based Systems
JF - Knowledge-Based Systems
M1 - 106675
ER -