Boosting Semi-Supervised Object Detection in Remote Sensing Images With Active Teaching
This work addresses the problem of object detection in remote sensing images for researchers and practitioners, offering an incremental improvement in annotation efficiency and detection performance.
The paper tackles the challenge of limited object-level annotations in remote sensing images by proposing a novel active learning method called SSOD-AT, which integrates a teacher-student network with an RoI comparison module and diversity criterion to boost semi-supervised object detection, achieving a 1% improvement over state-of-the-art methods on DOTA and DIOR datasets.
The lack of object-level annotations poses a significant challenge for object detection in remote sensing images (RSIs). To address this issue, active learning (AL) and semi-supervised learning (SSL) techniques have been proposed to enhance the quality and quantity of annotations. AL focuses on selecting the most informative samples for annotation, while SSL leverages the knowledge from unlabeled samples. In this letter, we propose a novel AL method to boost semi-supervised object detection (SSOD) for remote sensing images with a teacher student network, called SSOD-AT. The proposed method incorporates an RoI comparison module (RoICM) to generate high-confidence pseudo-labels for regions of interest (RoIs). Meanwhile, the RoICM is utilized to identify the top-K uncertain images. To reduce redundancy in the top-K uncertain images for human labeling, a diversity criterion is introduced based on object-level prototypes of different categories using both labeled and pseudo-labeled images. Extensive experiments on DOTA and DIOR, two popular datasets, demonstrate that our proposed method outperforms state-of-the-art methods for object detection in RSIs. Compared with the best performance in the SOTA methods, the proposed method achieves 1 percent improvement in most cases in the whole AL.