Probability Differential-Based Class Label Noise Purification for Object Detection in Aerial Images

Modern object detection for aerial images requires numerous annotated data. However, the data annotation process inevitably introduces noise due to the bird's eye view perspective of aerial images and the professional requirements of annotations. While recent noise-robust object detection methods achieved great success, the noise side effect during the early training stage was still a problem. As demonstrated in this letter, noise during the early training stage will cumulatively affect the final performance. Based on the abovementioned observations, we propose a training strategy called correction maximization training to purify the noisy annotations and then train models. In particular, we design a novel noise filter called the probability differential (PD) to identify and revise wrong labels. After purification, we train the detector with the revised dataset. Compared with the existing works, the proposed method could be adapted in most modern object detectors (e.g., Faster RCNN and RetinaNet) and requires little hyperparameter tuning across different datasets and models. Extensive experiments on DOTA show that the proposed method achieves the state-of-the-art results with both symmetric and asymmetric noise.