Pedestrian Similarity Extraction to Improve People Counting Accuracy

Xu Yang, Jose Gaspar, Wei Ke, Chan Lam, Yanwei Zheng, Weng Lou, Yapeng Wang

2019

Abstract

Current state-of-the-art single shot object detection pipelines, composed by an object detector such as Yolo, generate multiple detections for each object, requiring a post-processing Non-Maxima Suppression (NMS) algorithm to remove redundant detections. However, this pipeline struggles to achieve high accuracy, particularly in object counting applications, due to a trade-off between precision and recall rates. A higher NMS threshold results in fewer detections suppressed and, consequently, in a higher recall rate, as well as lower precision and accuracy. In this paper, we have explored a new pedestrian detection pipeline which is more flexible, able to adapt to different scenarios and with improved precision and accuracy. A higher NMS threshold is used to retain all true detections and achieve a high recall rate for different scenarios, and a Pedestrian Similarity Extraction (PSE) algorithm is used to remove redundant detentions, consequently improving counting accuracy. The PSE algorithm significantly reduces the detection accuracy volatility and its dependency on NMS thresholds, improving the mean detection accuracy for different input datasets.

Download


Paper Citation


in Harvard Style

Yang X., Gaspar J., Ke W., Lam C., Zheng Y., Lou W. and Wang Y. (2019). Pedestrian Similarity Extraction to Improve People Counting Accuracy.In Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-351-3, pages 548-555. DOI: 10.5220/0007381605480555


in Bibtex Style

@conference{icpram19,
author={Xu Yang and Jose Gaspar and Wei Ke and Chan Lam and Yanwei Zheng and Weng Lou and Yapeng Wang},
title={Pedestrian Similarity Extraction to Improve People Counting Accuracy},
booktitle={Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2019},
pages={548-555},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007381605480555},
isbn={978-989-758-351-3},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Pedestrian Similarity Extraction to Improve People Counting Accuracy
SN - 978-989-758-351-3
AU - Yang X.
AU - Gaspar J.
AU - Ke W.
AU - Lam C.
AU - Zheng Y.
AU - Lou W.
AU - Wang Y.
PY - 2019
SP - 548
EP - 555
DO - 10.5220/0007381605480555