Welcome to the new version of CaltechAUTHORS. Login is currently restricted to library staff. If you notice any issues, please email coda@library.caltech.edu
Published September 1, 2021 | Submitted
Report Open

Towards Reducing Labeling Cost in Deep Object Detection

Abstract

Deep neural networks have reached very high accuracy on object detection but their success hinges on large amounts of labeled data. To reduce the dependency on labels, various active-learning strategies have been proposed, typically based on the confidence of the detector. However, these methods are biased towards best-performing classes and can lead to acquired datasets that are not good representatives of the data in the testing set. In this work, we propose a unified framework for active learning, that considers both the uncertainty and the robustness of the detector, ensuring that the network performs accurately in all classes. Furthermore, our method is able to pseudo-label the very confident predictions, suppressing a potential distribution drift while further boosting the performance of the model. Experiments show that our method comprehensively outperforms a wide range of active-learning methods on PASCAL VOC07+12 and MS-COCO, having up to a 7.7% relative improvement, or up to 82% reduction in labeling cost.

Additional Information

Attribution 4.0 International (CC BY 4.0)

Attached Files

Submitted - 2106.11921.pdf

Files

2106.11921.pdf
Files (1.5 MB)
Name Size Download all
md5:f0522e0b37d775fe74296753bb2c343d
1.5 MB Preview Download

Additional details

Created:
August 20, 2023
Modified:
October 23, 2023