Improving Object Detection with Consistent Negative Sample Mining | |
Xiaolian Wang1,2; Xiyuan Hu1,2; Chen Chen1,2; Zhenfeng Fan1,2; Silong Peng1,2,3 | |
2019 | |
会议日期 | 2019 |
会议地点 | Sydney, Australia. |
英文摘要 | In object detection, training samples are divided into negatives and positives simply according to their initial positions on images. Samples which have low overlap with ground-truths are assigned to negatives, and positives otherwise. Once allocated, the negative and positive set are fixed in training. A usually overlooked issue is that certain negatives do not stick to their original states as training proceeds. They gradually regress towards foreground objects rather than away from them, which contradicts the nature of negatives. Training with such inconsistent negatives may confuse detectors in distinguishing between foreground and background, and thus makes training less effective. In this paper, we propose a consistent negative sample mining method to filter out biased negatives in training. Specifically, the neural network takes the regression performance into account, and dynamically activates consistent negatives which have both low input IoUs and low output IoUs for training. In the experiments, we evaluate our method on PASCAL VOC and KITTI datasets, and the improvements on both datasets demonstrate the effectiveness of our method. |
语种 | 英语 |
内容类型 | 会议论文 |
源URL | [http://ir.ia.ac.cn/handle/173211/25817] |
专题 | 自动化研究所_智能制造技术与系统研究中心_多维数据分析团队 自动化研究所_个人空间 |
通讯作者 | Chen Chen |
作者单位 | 1.Institute of Automation, Chinese Academy of Sciences, Beijing, China 2.University of Chinese Academy of Sciences, Beijing, China 3.Beijing ViSystem Corporation Limited, China |
推荐引用方式 GB/T 7714 | Xiaolian Wang,Xiyuan Hu,Chen Chen,et al. Improving Object Detection with Consistent Negative Sample Mining[C]. 见:. Sydney, Australia.. 2019. |
个性服务 |
查看访问统计 |
相关权益政策 |
暂无数据 |
收藏/分享 |
除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。
修改评论