TY - GEN
T1 - Self Supervised Learning for Multiple Object Tracking in 3D Point Clouds
AU - Kumar, Aakash
AU - Kini, Jyoti
AU - Mian, Ajmal
AU - Shah, Mubarak
N1 - Funding Information:
*This work was conducted at UCF and supported by Lockheed Martin Corporate Engineering, Technology and Operations (CETO) University Engagement (UE) - Research. Professor Ajmal Mian is the recipient of an Australian Research Council Future Fellowship Award (project number FT210100268) funded by the Australian Government.
Publisher Copyright:
© 2022 IEEE.
PY - 2022
Y1 - 2022
N2 - Multiple object tracking in 3D point clouds has applications in mobile robots and autonomous driving. This is a challenging problem due to the sparse nature of the point clouds and the added difficulty of annotation in 3D for supervised learning. To overcome these challenges, we propose a neural network architecture that learns effective object features and their affinities in a self supervised fashion for multiple object tracking in 3D point clouds captured with LiDAR sensors. For self supervision, we use two approaches. First, we generate two augmented LiDAR frames from a single real frame by applying translation, rotation and cutout to the objects. Second, we synthesize a LiDAR frame using CAD models or primitive geometric shapes and then apply the above three augmentations to them. Hence, the ground truth object locations and associations are known in both frames for self supervision. This removes the need to annotate object associations in real data, and additionally the need for training data collection and annotation for object detection in synthetic data. To the best of our knowledge, this is the first self supervised multiple object tracking method for 3D data. Our model achieves state of the art results.
AB - Multiple object tracking in 3D point clouds has applications in mobile robots and autonomous driving. This is a challenging problem due to the sparse nature of the point clouds and the added difficulty of annotation in 3D for supervised learning. To overcome these challenges, we propose a neural network architecture that learns effective object features and their affinities in a self supervised fashion for multiple object tracking in 3D point clouds captured with LiDAR sensors. For self supervision, we use two approaches. First, we generate two augmented LiDAR frames from a single real frame by applying translation, rotation and cutout to the objects. Second, we synthesize a LiDAR frame using CAD models or primitive geometric shapes and then apply the above three augmentations to them. Hence, the ground truth object locations and associations are known in both frames for self supervision. This removes the need to annotate object associations in real data, and additionally the need for training data collection and annotation for object detection in synthetic data. To the best of our knowledge, this is the first self supervised multiple object tracking method for 3D data. Our model achieves state of the art results.
UR - http://www.scopus.com/inward/record.url?scp=85146312200&partnerID=8YFLogxK
U2 - 10.1109/IROS47612.2022.9981793
DO - 10.1109/IROS47612.2022.9981793
M3 - Conference paper
AN - SCOPUS:85146312200
T3 - IEEE International Conference on Intelligent Robots and Systems
SP - 3754
EP - 3761
BT - IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2022
PB - IEEE, Institute of Electrical and Electronics Engineers
T2 - 2022 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 2022
Y2 - 23 October 2022 through 27 October 2022
ER -