A SSD-based Crowded Pedestrian Detection Method

Pedestrian detection has become a significant research topic in the field of computer vision. The performance of existing methods based on deep learning is not so good in pedestrian detection for complex background. Considering the problem of pedestrian detection in complex scenes with small and crowded objects, we propose a SSD-based crowded pedestrian detection method in this paper. Firstly, we increase density of default boxes on the horizontal direction by setting an offset, which can effectively eliminate the influence of missing matching default boxes and separate a person from the crowd much easier. So our detector is more suitable for complex scenes. Secondly, SSD is designed for general object detection, thus it is unfit for pedestrian detection because of the large aspect ratio of pedestrians. Therefore, we adopt abnormal 5*1 convolutional kernels instead of the standard 3*3 ones in order to adapt to pedestrian detection. Finally, we present experimental results on public benchmark datasets including Caltech dataset and INRIA dataset, which indicate that our method has better performance for pedestrian detection.

[1]  Trevor Darrell,et al.  Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation , 2013, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[2]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[3]  Bernt Schiele,et al.  New features and insights for pedestrian detection , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[4]  Gustavo Carneiro,et al.  Improving the performance of pedestrian detectors using convolutional learning , 2017, Pattern Recognit..

[5]  Kaiming He,et al.  Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks , 2015, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[6]  Bernt Schiele,et al.  Filtered channel features for pedestrian detection , 2015, 2015 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[7]  Ross B. Girshick,et al.  Fast R-CNN , 2015, 1504.08083.

[8]  Jian Sun,et al.  Spatial Pyramid Pooling in Deep Convolutional Networks for Visual Recognition , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[9]  Wei Liu,et al.  SSD: Single Shot MultiBox Detector , 2015, ECCV.

[10]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[11]  Bill Triggs,et al.  Histograms of oriented gradients for human detection , 2005, 2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05).

[12]  Pietro Perona,et al.  Pedestrian Detection: An Evaluation of the State of the Art , 2012, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[13]  Trevor Darrell,et al.  Caffe: Convolutional Architecture for Fast Feature Embedding , 2014, ACM Multimedia.

[14]  David A. McAllester,et al.  Object Detection with Discriminatively Trained Part Based Models , 2010, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[15]  Joon Hee Han,et al.  Local Decorrelation For Improved Pedestrian Detection , 2014, NIPS.

[16]  Shuicheng Yan,et al.  An HOG-LBP human detector with partial occlusion handling , 2009, 2009 IEEE 12th International Conference on Computer Vision.

[17]  Ali Farhadi,et al.  You Only Look Once: Unified, Real-Time Object Detection , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).