浏览全部资源
扫码关注微信
1.北京航空航天大学 仪器科学与光电工程学院 精密光机电一体化教育部重点实验室,北京 100191
2.季华实验室 智能机器人工程研究中心,广东 佛山 528200
[ "王 睿(1965-),女,北京人,博士,副教授,1987年、1990年于清华大学分别获得学士和硕士学位,2006年于北京航空航天大学获得博士学位,研究方向为信号测试与处理、图像处理、机器学习等。E-mail: wangr@buaa.edu.cn" ]
[ "温志庆(1964-),男,博士,正高级工程师,1994年于清华大学获得博士学位,研究方向为智能机器视觉、模式识别、3D成像、光学工程、人工智能等。E-mail: wenzq@jihualab.com" ]
收稿日期:2022-07-27,
修回日期:2022-09-11,
纸质出版日期:2023-07-10
移动端阅览
王睿,樊思杨,许婧文等.采用SVD协同训练的半监督实例级目标检测[J].光学精密工程,2023,31(13):2000-2007.
WANG Rui,FAN Siyang,XU Jingwen,et al.Semi-supervised instance object detection method based on SVD co-training[J].Optics and Precision Engineering,2023,31(13):2000-2007.
王睿,樊思杨,许婧文等.采用SVD协同训练的半监督实例级目标检测[J].光学精密工程,2023,31(13):2000-2007. DOI: 10.37188/OPE.20233113.2000.
WANG Rui,FAN Siyang,XU Jingwen,et al.Semi-supervised instance object detection method based on SVD co-training[J].Optics and Precision Engineering,2023,31(13):2000-2007. DOI: 10.37188/OPE.20233113.2000.
在室内实例物体目标检测中,传统深度学习需要大量人工标注的训练样本进行网络训练,费时费力,为此提出并实现了一种采用奇异值分解(Singular Value Decomposition,SVD)和协同训练的半监督实例级目标检测网络SVD-RCNN。挑选关键样本进行人工标注并预训练SVD-RCNN,以确保其获取更多先验知识,采用基于SVD的收敛-分解-微调策略,在SVD-RCNN中得到两个较强独立性的检测器以满足协同训练的要求,最后提出一种自适应的自标注策略,获得高质量的自标注及检测结果。在多个室内实例数据集上对该方法进行测试,在GMU数据集上只需人工标注199个样本,均值平均精度(mean Average Precision,mAP)达到了79.3%,相较于需标注3 851个样本的全监督Faster RCNN的81.3% mAP仅下降了2%。消融实验及系列实验证明了本文方法的有效性和普适性,本文提出的方法仅需人工标注5%的训练数据,即可达到与全监督学习相当的实例级目标检测精度,有利于智能机器人高效识别不同实例物体的实际应用。
Detecting indoor instance objects is useful for various applications. Traditional deep-learning methods require a large number of labeled samples for network training, making them time-consuming and labor-intensive. To address this problem, SVD-RCNN—a semi-supervised instance object detection network based on singular value decomposition (SVD) and co-training—is proposed. First, key samples are selected for manual labeling to pre-train SVD-RCNN, to ensure that it acquires more prior knowledge. Second, a convergence, decomposition, and finetuning strategy based on SVD is used to obtain two detectors with strong independence in SVD-RCNN to satisfy the requirements of co-training. Finally, an adaptive self-labeling strategy is used to obtain high-quality self-labeling and detection results. The method was tested on multiple indoor instance datasets. On the GMU dataset, it achieved a mean average precision of 79.3% with 199 manually labeled samples. This was only 2% lower than that (81.3%) of Faster RCNN with fully supervised learning, which required labeling 3 851 samples. Ablation studies and a series of experiments confirmed the effectiveness and universality of the method. The results indicated that the method only needs to manually label 5% of the training data to achieve instance-level detection accuracy comparable to that of fully supervised learning; thus, it is suitable for applications in which intelligent robots must efficiently identify different instance objects.
GRAUMAN K , LEIBE B . Visual object recognition [J]. Synthesis Lectures on Artificial Intelligence and Machine Learning , 2011 , 5 ( 2 ): 1 - 181 . doi: 10.2200/s00332ed1v01y201103aim011 http://dx.doi.org/10.2200/s00332ed1v01y201103aim011
范丽丽 , 赵宏伟 , 赵浩宇 , 等 . 基于深度卷积神经网络的目标检测研究综述 [J]. 光学 精密工程 , 2020 , 28 ( 5 ): 1152 - 1164 .
FAN L L , ZHAO H W , ZHAO H Y , et al . Survey of target detection based on deep convolutional neural networks [J]. Opt. Precision Eng. , 2020 , 28 ( 5 ): 1152 - 1164 . (in Chinese)
徐哲 , 耿杰 , 蒋雯 , 等 . 联合训练生成对抗网络的半监督分类方法 [J]. 光学 精密工程 , 2021 , 29 ( 5 ): 1127 - 1135 . doi: 10.37188/OPE.20212905.1127 http://dx.doi.org/10.37188/OPE.20212905.1127
XU ZH , GENG J , JIANG W , et al . Co-training generative adversarial networks for semi-supervised classification method [J]. Opt. Precision Eng. , 2021 , 29 ( 5 ): 1127 - 1135 . (in Chinese) . doi: 10.37188/OPE.20212905.1127 http://dx.doi.org/10.37188/OPE.20212905.1127
黄鸿 , 唐玉枭 , 段宇乐 . 半监督多图嵌入的高光谱影像特征提取 [J]. 光学 精密工程 , 2020 , 28 ( 2 ): 443 - 456 .
HUANG H , TANG Y X , DUAN Y L . Feature extraction of hyperspectral image with semi-supervised multi-graph embedding [J]. Opt. Precision Eng. , 2020 , 28 ( 2 ): 443 - 456 . (in Chinese)
JEONG J , VERMA V , HYUN M , et al . Interpolation-based semi-supervised learning for object detection [C]. 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 2025,2021 , Nashville, TN, USA. IEEE , 2021 : 11597 - 11606 . doi: 10.1109/cvpr46437.2021.01143 http://dx.doi.org/10.1109/cvpr46437.2021.01143
SOHN K , ZHANG Z , LI C , et al . A Simple Semi-supervised Learning Framework for Object Detection [EB/OL]. 2020 : arXiv : 2005 . 04757 . https://arxiv.org/abs/2005.04757 https://arxiv.org/abs/2005.04757 . doi: 10.48550/arXiv.2005.04757 http://dx.doi.org/10.48550/arXiv.2005.04757
LIU Y C , MA C Y , HE Z , et al . Unbiased Teacher for Semi-supervised Object Detection [EB/OL]. 2021 : arXiv : 2102 . 09480 . https://arxiv.org/abs/2102.09480 https://arxiv.org/abs/2102.09480 .
GUO Q S , MU Y , CHEN J Y , et al . Scale-equivalent distillation for semi-supervised object detection [C]. 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR). 1824,2022 , New Orleans, LA, USA. IEEE , 2022 : 14502 - 14511 . doi: 10.1109/cvpr52688.2022.01412 http://dx.doi.org/10.1109/cvpr52688.2022.01412
LI G , LI X , WANG Y J , et al . PseCo : Pseudo Labeling and Consistency Training for Semi-supervised Object Detection [M]. Lecture Notes in Computer Science . Cham : Springer Nature Switzerland , 2022 : 457 - 472 . doi: 10.1007/978-3-031-20077-9_27 http://dx.doi.org/10.1007/978-3-031-20077-9_27
NING X , WANG X R , XU S H , et al . A review of research on co-training [J]. Concurrency and Computation: Practice and Experience , 2021 : doi: 10.1002/cpe.6276 http://dx.doi.org/10.1002/cpe.6276 .
DAN K . A singularly valuable decomposition: the SVD of a matrix [J]. The College Mathematics Journal , 1996 , 27 ( 1 ): 2 - 23 . doi: 10.1080/07468342.1996.11973744 http://dx.doi.org/10.1080/07468342.1996.11973744
REN S Q , HE K M , GIRSHICK R , et al . Faster R-CNN: towards real-time object detection with region proposal networks [J]. IEEE Transactions on Pattern Analysis and Machine Intelligence , 2017 , 39 ( 6 ): 1137 - 1149 . doi: 10.1109/tpami.2016.2577031 http://dx.doi.org/10.1109/tpami.2016.2577031
RUI W , YING L . Real-time 3D object detection in unstructured environments [C]. 2017 First International Conference on Electronics Instrumentation & Information Systems (EIIS). 35,2017 , Harbin, China. IEEE , 2018 : 1 - 6 . doi: 10.1109/eiis.2017.8298584 http://dx.doi.org/10.1109/eiis.2017.8298584
VERDU S . Fifty years of Shannon theory [J]. IEEE Transactions on Information Theory , 1998 , 44 ( 6 ): 2057 - 2078 . doi: 10.1109/18.720531 http://dx.doi.org/10.1109/18.720531
GATYS L A , ECKER A S , BETHGE M . Image style transfer using convolutional neural networks [C]. 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR). 2730,2016 , Las Vegas, NV, USA. IEEE , 2016 : 2414 - 2423 . doi: 10.1109/cvpr.2016.265 http://dx.doi.org/10.1109/cvpr.2016.265
EVERINGHAM M , GOOL L , WILLIAMS C K , et al . The pascal visual object classes (VOC) challenge [J]. International Journal of Computer Vision , 2010 , 88 ( 2 ): 303 - 338 . doi: 10.1007/s11263-009-0275-4 http://dx.doi.org/10.1007/s11263-009-0275-4
GEORGAKIS G , ALIMOOR REZA M , MOUSAVIAN A , et al . Multiview RGB-D dataset for object instance detection [C]. 2016 Fourth International Conference on 3D Vision (3DV). 2528,2016 , Stanford, CA, USA. IEEE , 2016 : 426 - 434 . doi: 10.1109/3dv.2016.52 http://dx.doi.org/10.1109/3dv.2016.52
AMMIRATO P , POIRSON P , PARK E , et al . A dataset for developing and benchmarking active vision [C]. 2017 IEEE International Conference on Robotics and Automation (ICRA). May 29 - June 3 , 2017 , Singapore. IEEE , 2017: 1378 - 1385 . doi: 10.1109/icra.2017.7989164 http://dx.doi.org/10.1109/icra.2017.7989164
XU M D , ZHANG Z , HU H , et al . End-to-end semi-supervised object detection with soft teacher [C]. 2021 IEEE/CVF International Conference on Computer Vision (ICCV). 1017,2021 , Montreal, QC, Canada. IEEE , 2022 : 3040 - 3049 . doi: 10.1109/iccv48922.2021.00305 http://dx.doi.org/10.1109/iccv48922.2021.00305
0
浏览量
204
下载量
0
CSCD
关联资源
相关文章
相关作者
相关机构