ABSTRACT
The occlusion in crowded scenes and the interference of similar objects in the background are one of the main reasons that lead to missed pedestrian detection. In response to this problem, an improved Cascade R-CNN pedestrian detection algorithm using dynamic regressors is proposed. Firstly, the offset of each sample is adjusted to a dynamic offset and normalization to strengthen the regression performance of each regressor. Then, based on the preliminary regression, a secondary detection module of occluded pedestrians is constructed to further enhance the discrimination between pedestrians and pedestrians. The proposed algorithm performed ablation experiments on the datasets Caltech, CityPersons and CrowdHuman, and the missed detection rates were reduced by 21.3%, 5.0% and 8.4% on the Heavy subset of the above three datasets, respectively. The experimental results show that the improved Cascade R-CNN algorithm is strongly robust to obstructed pedestrians.
- HASAN I, LIAO S, LI J, Pedestrian detection: The elephant in the room. arXiv preprint arXiv:200308799, 2020.Google Scholar
- ZHANG S, BENENSON R, OMRAN M, How far are we from solving pedestrian detection?; proceedings of the Proceedings of the iEEE conference on computer vision and pattern recognition, F, 2016 [C].Google ScholarCross Ref
- WANG X, XIAO T, JIANG Y, Repulsion loss: Detecting pedestrians in a crowd; proceedings of the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, F, 2018 [C].Google Scholar
- LIU S, HUANG D, WANG Y. Adaptive nms: Refining pedestrian detection in a crowd; proceedings of the Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, F, 2019 [C].Google Scholar
- ZHANG S, WEN L, BIAN X, Occlusion-aware R-CNN: detecting pedestrians in a crowd; proceedings of the Proceedings of the European Conference on Computer Vision (ECCV), F, 2018 [C].Google Scholar
- CAI Z, VASCONCELOS N. Cascade r-cnn: Delving into high quality object detection; proceedings of the Proceedings of the IEEE conference on computer vision and pattern recognition, F, 2018 [C].Google Scholar
- ZHANG L, LIN L, LIANG X, Is faster R-CNN doing well for pedestrian detection?; proceedings of the European conference on computer vision, F, 2016 [C]. Springer.Google ScholarCross Ref
- VIOLA P, JONES M. Rapid object detection using a boosted cascade of simple features; proceedings of the Proceedings of the 2001 IEEE computer society conference on computer vision and pattern recognition CVPR 2001, F, 2001 [C]. IEEE.Google ScholarCross Ref
- MAO J, XIAO T, JIANG Y, What can help pedestrian detection?; proceedings of the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, F, 2017 [C].Google ScholarCross Ref
- JIANG B, LUO R, MAO J, Acquisition of localization confidence for accurate object detection; proceedings of the Proceedings of the European conference on computer vision (ECCV), F, 2018 [C].Google ScholarDigital Library
- WU Y, CHEN Y, YUAN L, Rethinking classification and localization for object detection; proceedings of the Proceedings of the IEEE/CVF conference on computer vision and pattern recognition, F, 2020 [C].Google ScholarCross Ref
- PAPAGEORGIOU C P, OREN M, POGGIO T. A general framework for object detection; proceedings of the Sixth International Conference on Computer Vision (IEEE Cat No 98CH36271), F, 1998 [C]. IEEE.Google ScholarCross Ref
- CHEN K, WANG J, PANG J, MMDetection: Open mmlab detection toolbox and benchmark . arXiv preprint arXiv:190607155, 2019.Google Scholar
- CAO Y, CHEN K, LOY C C, Prime sample attention in object detection; proceedings of the Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition, F, 2020 [C].Google ScholarCross Ref
- CHEN J, LIU D, XU T, Is sampling heuristics necessary in training deep object detectors? [J]. arXiv preprint arXiv:190904868, 2019.Google Scholar
- DOLLáR P, WOJEK C, SCHIELE B, Pedestrian detection: A benchmark; proceedings of the 2009 IEEE Conference on Computer Vision and Pattern Recognition, F, 2009 [C]. IEEE.Google Scholar
- ZHANG S, BENENSON R, SCHIELE B. Citypersons: A diverse dataset for pedestrian detection; proceedings of the Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, F, 2017 [C].Google Scholar
- SHAO S, ZHAO Z, LI B, Crowdhuman: A benchmark for detecting human in a crowd. arXiv preprint arXiv:180500123, 2018.Google Scholar
- HE K, ZHANG X, REN S, Deep residual learning for image recognition; proceedings of the Proceedings of the IEEE conference on computer vision and pattern recognition, F, 2016 [C].Google ScholarCross Ref
- XIE S, GIRSHICK R, DOLLáR P, Aggregated residual transformations for deep neural networks; proceedings of the Proceedings of the IEEE conference on computer vision and pattern recognition, F, 2017 [C].Google ScholarCross Ref
Recommendations
Multimodal pedestrian detection using metaheuristics with deep convolutional neural network in crowded scenes
AbstractPedestrian detection (PD) is a vital computer vision (CV) problem that is highly employed in several real-time applications, namely autonomous driving methods, robotics, and security observing methods. Simulated by deep learning (DL) approaches ...
Highlights- Develop an automated pedestrian detection in crowded scenes using the MMPD-MDCNN technique.
- Study multimodal ensemble learning process for pedestrian detection.
- For tuning deep learning models, the HHO algorithm is applied.
- ...
A New Pedestrian Detect Method in Crowded Scenes
GREENCOM-ITHINGS-CPSCOM '13: Proceedings of the 2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical and Social ComputingMost existing pedestrian detection methods always focus on improving detect accuracy of single pedestrian detection, but in this paper we focus on detect crowded pedestrians and recognizing adjacent or overlapped pedestrian exactly. We pro-pose a ...
Real-time high-precision pedestrian tracking: a detection–tracking–correction strategy based on improved SSD and Cascade R-CNN
AbstractThe existing pedestrian tracking applications are challenging to balance real-time performance and accuracy. We propose a detection–tracking–correction strategy based on the improved single-shot multi-box detector (SSD), Deep-SORT, and the ...
Comments