摘要
针对像素级自适应较大的图像翻译偏差,特征级自适应的源偏判别风险以及弱监督学习无法兼顾检测准确性和实时性等问题,提出了多元化域移位器和伪边界框生成器以逐步调整预训练模型,在像素级与特征级渐进完成自适应的域迁移框架。通过域移位器从源域生成多样化的中间域图像调整检测模型以弥合域差距,减小图像翻译偏差。将中间域作为监督的源域,并结合目标域中的图像级标签生成伪标注图像调整检测模型以改善源偏判别性。基于SSD算法构建与域迁移框架相匹配的实时目标检测器,实现弱监督条件下的实时目标检测。在PASCAL VOC迁移至Clipart1k等数据集上的mAP优于现有方法0.4%~4.7%,检测速度为32 FPS~47 FPS,提高准确率的同时满足了实时检测的要求,具有更优越的迁移检测性能。
Aiming at the problems of large image translation bias at the pixel-level adaptation,the risk of source-bias discrimination at the feature-level adaptation,and the inability of weakly supervised learning to balance detection accuracy and realtime performance,a diversified domain shifter and pseudo bounding box generator are proposed to gradually adjust the pre-training model.The adaptive cross-domain framework is gradually completed at pixel-level and feature-level.A diversified intermediate domain adjustment detection model is generated from the source domain by a domain shifter to bridge the domain gap and reduce the image translation bias.The intermediate domain is used as the supervised source domain,and the pseudo-labeled image adjustment detection model is generated by combining image-level annotations in the target domain to improve source-bias discrimination.A real-time object detector matching the cross-domain framework is constructed based on SSD algorithm to realize real-time object detection under weakly supervised conditions.The mAP on PASCAL VOC migrated to Clipart1k and other datasets is 0.4%~4.7%better than the existing methods.The detection speed is 32 FPS~47 FPS.This improves the accuracy and meets the requirements of real-time detection,and has better migration detection performance.
作者
李成严
郑企森
王昊
LI Chengyan;ZHENG Qisen;WANG Hao(School of Computer Science and Technology,Harbin University of Science and Technology,Harbin 150080,China)
出处
《哈尔滨理工大学学报》
CAS
北大核心
2024年第3期11-19,共9页
Journal of Harbin University of Science and Technology
基金
国家自然科学基金(61772160)
黑龙江省自然科学基金(LH2021F032)
黑龙江省重点研发计划项目(2022ZX01A34)。
关键词
实时目标检测
弱监督学习
域自适应
图像翻译网络
SSD算法
real-time object detection
weakly supervised learning
domain adaptation
image translation network
SSD algorithm
