Localization-based active learning (LOCAL) for object detection in 3D point clouds

Aimee Moses; Srikanth Jakkampudi; Cheryl Danner; Derek Biega

doi:10.1117/12.2618513

27 May 2022 Localization-based active learning (LOCAL) for object detection in 3D point clouds

Aimee Moses, Srikanth Jakkampudi, Cheryl Danner, Derek Biega

Proceedings Volume 12099, Geospatial Informatics XII; 1209907 (2022) https://doi.org/10.1117/12.2618513
Event: SPIE Defense + Commercial Sensing, 2022, Orlando, Florida, United States

Abstract

Deep learning-based object detection and classification in 3D point clouds has numerous applications including defense, autonomous driving, and augmented reality. A challenge in applying deep learning to point clouds is the frequent scarcity of labeled data. Often, one must manually label a large quantity of data for the model to be useful in application. To overcome this challenge, active learning provides a means of minimizing the manual labeling required. The crux of active learning algorithms is defining and calculating the potential added “value” of labeling each unlabeled sample. We introduce a novel active learning algorithm, LOCAL, with an anchorbased object detection architecture, a modified object matching strategy, and an acquisition metric designed for object detection in any dimension. We compare the performance of common acquisition functions to our novel metric that utilizes all of the model outputs—including both bounding box localizations and softmax classification scores—to capture both the classification and spatial uncertainty in the model. Finally, we identify opportunities for further exploration, such as alternative measures of spatial uncertainty as well as increasing the stochasticity of the model in order to improve robustness of the algorithm.

Conference Presentation

Citation Download Citation

Aimee Moses, Srikanth Jakkampudi, Cheryl Danner, and Derek Biega "Localization-based active learning (LOCAL) for object detection in 3D point clouds", Proc. SPIE 12099, Geospatial Informatics XII, 1209907 (27 May 2022); https://doi.org/10.1117/12.2618513

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available