Authors:
Xu Yang
1
;
Jose Gaspar
1
;
Wei Ke
1
;
Chan Tong Lam
1
;
Yanwei Zheng
2
;
Weng Hong Lou
1
and
Yapeng Wang
3
Affiliations:
1
School of Public Administration, Macao Polytechnic Institute, Macao S.A.R and China
;
2
Beihang University and China
;
3
Information Systems Research Centre, Macao Polytechnic Institute, Macao S.A.R and China
Keyword(s):
Pedestrian Detection and Counting, Pedestrian Similarity Extraction, Non-Maxima Suppression (NMS), Yolo, Convolutional Neural Networks (CNN).
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 alg
orithm significantly reduces the detection accuracy volatility and its dependency on NMS thresholds, improving the mean detection accuracy for different input datasets.
(More)