Accurate Single Stage Detector Using Recurrent Rolling Convolution

被引:139
作者
Ren, Jimmy [1 ]
Chen, Xiaohao [1 ]
Liu, Jianbo [1 ]
Sun, Wenxiu [1 ]
Pang, Jiahao [1 ]
Yan, Qiong [1 ]
Tai, Yu-Wing [1 ]
Xu, Li [1 ]
机构
[1] SenseTime Grp Ltd, Hong Kong, Hong Kong, Peoples R China
来源
30TH IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR 2017) | 2017年
关键词
D O I
10.1109/CVPR.2017.87
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
Most of the recent successful methods in accurate object detection and localization used some variants of R-CNN style two stage Convolutional Neural Networks (CNN) where plausible regions were proposed in the first stage then followed by a second stage for decision refinement. Despite the simplicity of training and the efficiency in deployment, the single stage detection methods have not been as competitive when evaluated in benchmarks consider mAP for high IoU thresholds. In this paper, we proposed a novel single stage end-to-end trainable object detection network to overcome this limitation. We achieved this by introducing Recurrent Rolling Convolution (RRC) architecture over multi-scale feature maps to construct object classifiers and bounding box regressors which are "deep in context". We evaluated our method in the challenging KITTI dataset which measures methods under IoU threshold of 0.7. We showed that with RRC, a single reduced VGG-16 based model already significantly outperformed all the previously published results. At the time this paper was written our models ranked the first in KITTI car detection (the hard level), the first in cyclist detection and the second in pedestrian detection. These results were not reached by the previous single stage methods. The code is publicly available.
引用
收藏
页码:752 / 760
页数:9
相关论文
共 23 条
[1]  
[Anonymous], 2016, CVPR
[2]  
Bahdanau D, 2016, Arxiv, DOI [arXiv:1409.0473, DOI 10.48550/ARXIV.1409.0473]
[3]   A Unified Multi-scale Deep Convolutional Neural Network for Fast Object Detection [J].
Cai, Zhaowei ;
Fan, Quanfu ;
Feris, Rogerio S. ;
Vasconcelos, Nuno .
COMPUTER VISION - ECCV 2016, PT IV, 2016, 9908 :354-370
[4]   Monocular 3D Object Detection for Autonomous Driving [J].
Chen, Xiaozhi ;
Kundu, Kaustav ;
Zhang, Ziyu ;
Ma, Huimin ;
Fidler, Sanja ;
Urtasun, Raquel .
2016 IEEE CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR), 2016, :2147-2156
[5]  
Chen XZ, 2015, ADV NEUR IN, V28
[6]  
Dai J., 2016, ADV NEURAL INFORM PR, V29, P379, DOI [DOI 10.1016/J.JPOWSOUR.2007.02.075, DOI 10.48550/ARXIV.1605.06409, DOI 10.1109/CVPR.2017.690]
[7]  
Geiger Andreas, 2012, C COMP VIS PATT REC, P2
[8]  
Gidaris S., 2016, P IEEE C COMP VIS PA
[9]  
Girshick, 2015, P IEEE INT C COMP VI, DOI [10.1109/ICCV.2015.169, DOI 10.1109/ICCV.2015.169]
[10]  
Girshick R., 2014, IEEE C COMP VIS PATT, DOI [DOI 10.1109/CVPR.2014.81, 10.1109/CVPR.2014.81]