Incremental Object Detection Method Based on Border Distance Measurement

• Published in: Ji suan ji ke xue Vol. 49; no. 8; pp. 136 - 142
• Main Authors: Liu, Dong-mei, Xu, Yang, Wu, Ze-bin, Liu, Qian, Song, Bin, Wei, Zhi-hui
• Format: Journal Article
• Language: Chinese
• Published: Chongqing Guojia Kexue Jishu Bu 01.08.2022; Editorial office of Computer Science
• Subjects: Accuracy; Algorithms; Annotations; Datasets; Detectors; Distance measurement; Distillation; Feature extraction; Image classification; Labels; Learning; Model accuracy; Modules; object detection|label selection|incremental learning|attention module|catastrophic forgetting|pseudo label; Object recognition
• ISSN: 1002-137X
• DOI: 10.11896/jsjkx.220100132

Incremental learning has achieved good results in image classification, but it is challenging to apply incremental learning to multi-class object detection.Object detection is more complex than image classification, which combines classification and border regression.At present, the most advanced incremental object detectors adopt the external fixed region suggestion method based on knowledge distillation, which consumes a lot of time and cost.For single-stage detectors, due to the lack of annotation and region advice information for the old class, old objects are usually identified by the detector as the background, resulting in catastrophic forgetting.In this paper, a label selection algorithm based on border distance metric is proposed.It uses the detection results of the old model and the existing dataset labels to select and merge by measuring the coincidence of the bounding boxes, making up for the lack of annotations of the old objects in the new dataset and alleviating catastrophic forgetting.In addit