TY - GEN
T1 - SANet
T2 - 2019 International Conference on Biometrics, ICB 2019
AU - Shi, Lei
AU - Xu, Xiang
AU - Kakadiaris, Ioannis A.
N1 - Funding Information:
This material is based upon work supported by the U.S. Department of Homeland Security under Grant Award Number 2017-ST-BTI-0001-0201. This grant is awarded to the Borders, Trade, and Immigration (BTI) Institute: A DHS Center of Excellence led by the University of Houston, and includes support for the project EDGE awarded to the University of Houston.
Publisher Copyright:
© 2019 IEEE.
PY - 2019/6
Y1 - 2019/6
N2 - Recently, significant effort has been devoted to exploring the role of feature fusion and enriching contextual information on detecting multi-scale faces. However, simply integrating features of different levels could lead to introducing significant noise. Moreover, recently proposed approaches of enriching contextual information are not efficient or ignore the gridding artifacts produced by dilated convolution. To tackle these issues, we developed a smoothed attention network (dubbed SANet), which introduces an Attention-guided Feature Fusion Module (AFFM) and a Smoothed Context Enhancement Module (SCEM). In particular, the AFFM applies an attention module to high-level semantic features and fuses attention-focused features with low-level semantic features to reduce the noise of the fused feature map. The SCEM stacks dilated convolution and convolution layers alternately to re-learn the relationship among completely separate sets of units produced by dilated convolution to maintain consistency of local information. The SANet achieves promising results on the WIDER FACE validation and testing datasets and is state-of-the-art on the UFDD dataset.
AB - Recently, significant effort has been devoted to exploring the role of feature fusion and enriching contextual information on detecting multi-scale faces. However, simply integrating features of different levels could lead to introducing significant noise. Moreover, recently proposed approaches of enriching contextual information are not efficient or ignore the gridding artifacts produced by dilated convolution. To tackle these issues, we developed a smoothed attention network (dubbed SANet), which introduces an Attention-guided Feature Fusion Module (AFFM) and a Smoothed Context Enhancement Module (SCEM). In particular, the AFFM applies an attention module to high-level semantic features and fuses attention-focused features with low-level semantic features to reduce the noise of the fused feature map. The SCEM stacks dilated convolution and convolution layers alternately to re-learn the relationship among completely separate sets of units produced by dilated convolution to maintain consistency of local information. The SANet achieves promising results on the WIDER FACE validation and testing datasets and is state-of-the-art on the UFDD dataset.
UR - http://www.scopus.com/inward/record.url?scp=85081061629&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85081061629&partnerID=8YFLogxK
U2 - 10.1109/ICB45273.2019.8987285
DO - 10.1109/ICB45273.2019.8987285
M3 - Conference contribution
AN - SCOPUS:85081061629
T3 - 2019 International Conference on Biometrics, ICB 2019
BT - 2019 International Conference on Biometrics, ICB 2019
PB - Institute of Electrical and Electronics Engineers Inc.
Y2 - 4 June 2019 through 7 June 2019
ER -