Employing feature mixture for active learning of object detection

Zhang，Licheng1; Lam，Siew Kei2; Luo，Dingsheng3; Wu，Xihong3

2024-08-14

10.1016/j.neucom.2024.127883

Neurocomputing   影响因子和分区

0925-2312

1872-8286

Active learning aims to select the most informative samples for annotation from a large amount of unlabeled data, in order to reduce time-consuming and labor-intensive manual labeling efforts. Although active learning for object detection has made substantial progress in recent years, developing an accurate and efficient active learning algorithm for object detection remains a challenge. In this paper, we propose a novel unsupervised active learning method for deep object detection. This is based on our hypotheses that an object is more likely to be wrongly predicted by the model, if the prediction changes when its feature representations are slightly mixed by another feature representations at a very small ratio. Such unlabeled samples can be regarded as informative samples that can be selected by active learning. Our method employs base representations of all categories generated from the object detection network to examine the robustness of every detected object. We design a scoring function to calculate the informative score of each unlabeled image. We conduct extensive experiments on two public datasets, i.e., PASCAL VOC and MS-COCO. Experiment results show that our approach consistently outperforms state-of-the-art single-model based methods with significant margins. Our approach also performs on par with multi-model based methods, at much lesser computational cost.

Active learning Feature mixture Object detection SSD

SCI ; EI

英语

第一 ; 通讯

20242116138193

Deep learning ; Learning algorithms ; Learning systems ; Object recognition

Ergonomics and Human Factors Engineering:461.4 ; Data Processing and Image Processing:723.2 ; Machine Learning:723.4.2

COMPUTER SCIENCE

2-s2.0-85193823347

Scopus

1.Southern University of Science and Technology,Shenzhen,Guangdong,518055,China
2.Nanyang Technological University,639798,Singapore
3.Peking University,Beijing,100871,China

Zhang，Licheng,Lam，Siew Kei,Luo，Dingsheng,et al. Employing feature mixture for active learning of object detection[J]. Neurocomputing,2024,594.

Zhang，Licheng,Lam，Siew Kei,Luo，Dingsheng,&Wu，Xihong.(2024).Employing feature mixture for active learning of object detection.Neurocomputing,594.

Zhang，Licheng,et al."Employing feature mixture for active learning of object detection".Neurocomputing 594(2024).