TY - GEN
T1 - Sparse activation maps for interpreting 3D object detection
AU - Chen, Qiuxiao
AU - Li, Pengfei
AU - Xu, Meng
AU - Qi, Xiaojun
N1 - Publisher Copyright:
© 2021 IEEE.
PY - 2021/6
Y1 - 2021/6
N2 - We propose a technique to generate "visual explanations"for interpretability of volumetric-based 3D object detection networks. Specifically, we use the average pooling of weights to produce a Sparse Activation Map (SAM) which highlights the important regions of the 3D point cloud data. The SAMs is applicable to any volumetric-based models (model agnostic) to provide intuitive intermediate results at different layers to understand the complex network structures. The SAMs at the 3D feature map layer and the 2D feature map layer help to understand the effectiveness of neurons to capture the object information. The SAMs at the classification layer for each object class helps to understand the true positives and false positives of each network. The experimental results on the KITTI dataset demonstrate the visual observations of the SAM match the detection results of three volumetric-based models.
AB - We propose a technique to generate "visual explanations"for interpretability of volumetric-based 3D object detection networks. Specifically, we use the average pooling of weights to produce a Sparse Activation Map (SAM) which highlights the important regions of the 3D point cloud data. The SAMs is applicable to any volumetric-based models (model agnostic) to provide intuitive intermediate results at different layers to understand the complex network structures. The SAMs at the 3D feature map layer and the 2D feature map layer help to understand the effectiveness of neurons to capture the object information. The SAMs at the classification layer for each object class helps to understand the true positives and false positives of each network. The experimental results on the KITTI dataset demonstrate the visual observations of the SAM match the detection results of three volumetric-based models.
UR - http://www.scopus.com/inward/record.url?scp=85116020659&partnerID=8YFLogxK
U2 - 10.1109/CVPRW53098.2021.00017
DO - 10.1109/CVPRW53098.2021.00017
M3 - Conference contribution
AN - SCOPUS:85116020659
T3 - IEEE Computer Society Conference on Computer Vision and Pattern Recognition Workshops
SP - 76
EP - 84
BT - Proceedings - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
PB - IEEE Computer Society
T2 - 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, CVPRW 2021
Y2 - 19 June 2021 through 25 June 2021
ER -