TY - JOUR
T1 - Boost 3-D Object Detection via Point Clouds Segmentation and Fused 3-D GIoU-L Loss
AU - Chen, Yaran
AU - Li, Haoran
AU - Gao, Ruiyuan
AU - Zhao, Dongbin
N1 - Publisher Copyright:
© 2012 IEEE.
PY - 2022/2/1
Y1 - 2022/2/1
N2 - The 3-D object detection is crucial for many real-world applications, attracting many researchers' attention. Beyond 2-D object detection, 3-D object detection usually needs to extract appearance, depth, position, and orientation information from light detection and ranging (LiDAR) and camera sensors. However, due to more degrees of freedom and vertices, existing detection methods that directly transform from 2-D to 3-D still face several challenges, such as exploding increase of anchors' number and inefficient or hard-to-optimize objective. To this end, we present a fast segmentation method for 3-D point clouds to reduce anchors, which can largely decrease the computing cost. Moreover, taking advantage of 3-D generalized Intersection of Union (GIoU) and L1 losses, we propose a fused loss to facilitate the optimization of 3-D object detection. A series of experiments show that the proposed method has alleviated the abovementioned issues effectively.
AB - The 3-D object detection is crucial for many real-world applications, attracting many researchers' attention. Beyond 2-D object detection, 3-D object detection usually needs to extract appearance, depth, position, and orientation information from light detection and ranging (LiDAR) and camera sensors. However, due to more degrees of freedom and vertices, existing detection methods that directly transform from 2-D to 3-D still face several challenges, such as exploding increase of anchors' number and inefficient or hard-to-optimize objective. To this end, we present a fast segmentation method for 3-D point clouds to reduce anchors, which can largely decrease the computing cost. Moreover, taking advantage of 3-D generalized Intersection of Union (GIoU) and L1 losses, we propose a fused loss to facilitate the optimization of 3-D object detection. A series of experiments show that the proposed method has alleviated the abovementioned issues effectively.
KW - 3-D object detection
KW - generalized Intersection of Union (GIoU) loss
KW - segmentation
UR - http://www.scopus.com/inward/record.url?scp=85124053234&partnerID=8YFLogxK
U2 - 10.1109/TNNLS.2020.3028964
DO - 10.1109/TNNLS.2020.3028964
M3 - Article
C2 - 33112754
AN - SCOPUS:85124053234
SN - 2162-237X
VL - 33
SP - 762
EP - 773
JO - IEEE Transactions on Neural Networks and Learning Systems
JF - IEEE Transactions on Neural Networks and Learning Systems
IS - 2
ER -