TY - GEN
T1 - Two-stream network for online quadrotor detection without dedicated annotations
AU - Jin, Ruibing
AU - Wang, Jianliang
N1 - Publisher Copyright:
© 2019 IEEE.
PY - 2019/7
Y1 - 2019/7
N2 - Aerial vehicles detection, especially quadrotor detection, is important for cooperative unmanned aerial vehicles (UAV). However, traditional object detection approaches require a well-annotated and large scale dataset which is labor-intensive and time-consuming. Since the appearance of UAV is various, it is difficult to establish a dataset covering all kinds of UAVs. An intuitive solution for it is to train a network on an off-the-shelf dataset of a class which is under the same parent category as quadrotors. However, domain gap between these two classes hinders the performance of the trained network. To address this issue, a two-stream network is proposed. The appearance information (from a spatial stream) and the motion information (from a temporal stream) are incorporated in this network. A fusion module, cross proposal module, is proposed to fuse these two streams. To verify the performance of this two-stream network, a fully annotated dataset of quadrotors is established. Extensive experiments are conducted on it and the results show that our two-stream network performs better than traditional detection approaches in this task.
AB - Aerial vehicles detection, especially quadrotor detection, is important for cooperative unmanned aerial vehicles (UAV). However, traditional object detection approaches require a well-annotated and large scale dataset which is labor-intensive and time-consuming. Since the appearance of UAV is various, it is difficult to establish a dataset covering all kinds of UAVs. An intuitive solution for it is to train a network on an off-the-shelf dataset of a class which is under the same parent category as quadrotors. However, domain gap between these two classes hinders the performance of the trained network. To address this issue, a two-stream network is proposed. The appearance information (from a spatial stream) and the motion information (from a temporal stream) are incorporated in this network. A fusion module, cross proposal module, is proposed to fuse these two streams. To verify the performance of this two-stream network, a fully annotated dataset of quadrotors is established. Extensive experiments are conducted on it and the results show that our two-stream network performs better than traditional detection approaches in this task.
UR - http://www.scopus.com/inward/record.url?scp=85075785069&partnerID=8YFLogxK
U2 - 10.1109/ICCA.2019.8899924
DO - 10.1109/ICCA.2019.8899924
M3 - Conference contribution
AN - SCOPUS:85075785069
T3 - IEEE International Conference on Control and Automation, ICCA
SP - 561
EP - 566
BT - 2019 IEEE 15th International Conference on Control and Automation, ICCA 2019
PB - IEEE Computer Society
T2 - 15th IEEE International Conference on Control and Automation, ICCA 2019
Y2 - 16 July 2019 through 19 July 2019
ER -