TY - GEN
T1 - Score-CAM
T2 - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020
AU - Wang, Haofan
AU - Wang, Zifan
AU - Du, Mengnan
AU - Yang, Fan
AU - Zhang, Zijian
AU - Ding, Sirui
AU - Mardziel, Piotr
AU - Hu, Xia
N1 - Publisher Copyright:
© 2020 IEEE.
PY - 2020/6
Y1 - 2020/6
N2 - Recently, increasing attention has been drawn to the internal mechanisms of convolutional neural networks, and the reason why the network makes specific decisions. In this paper, we develop a novel post-hoc visual explanation method called Score-CAM based on class activation mapping. Unlike previous class activation mapping based approaches, Score-CAM gets rid of the dependence on gradients by obtaining the weight of each activation map through its forward passing score on target class, the final result is obtained by a linear combination of weights and activation maps. We demonstrate that Score-CAM achieves better visual performance and fairness for interpreting the decision making process. Our approach outperforms previous methods on both recognition and localization tasks, it also passes the sanity check. We also indicate its application as debugging tools. The implementation is available1.
AB - Recently, increasing attention has been drawn to the internal mechanisms of convolutional neural networks, and the reason why the network makes specific decisions. In this paper, we develop a novel post-hoc visual explanation method called Score-CAM based on class activation mapping. Unlike previous class activation mapping based approaches, Score-CAM gets rid of the dependence on gradients by obtaining the weight of each activation map through its forward passing score on target class, the final result is obtained by a linear combination of weights and activation maps. We demonstrate that Score-CAM achieves better visual performance and fairness for interpreting the decision making process. Our approach outperforms previous methods on both recognition and localization tasks, it also passes the sanity check. We also indicate its application as debugging tools. The implementation is available1.
UR - http://www.scopus.com/inward/record.url?scp=85090115581&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85090115581&partnerID=8YFLogxK
U2 - 10.1109/CVPRW50498.2020.00020
DO - 10.1109/CVPRW50498.2020.00020
M3 - Conference contribution
AN - SCOPUS:85090115581
T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
SP - 111
EP - 119
BT - Proceedings - 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2020
PB - IEEE Computer Society
Y2 - 14 June 2020 through 19 June 2020
ER -