High-efficiency face detection and tracking method for numerous pedestrians through face candidate generation
- PDF / 4,852,313 Bytes
- 26 Pages / 439.37 x 666.142 pts Page_size
- 10 Downloads / 235 Views
High-efficiency face detection and tracking method for numerous pedestrians through face candidate generation Deng-Yuan Huang 1 & Chao-Ho Chen 2 & Tsong-Yi Chen 2 & Wu-Chih Hu 3 & Zhi-Bin Guo 2 & Cheng-Kang Wen 4 Received: 9 September 2019 / Revised: 20 August 2020 / Accepted: 28 August 2020 # Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract
This paper is dedicated to developing high-efficiency face detection and tracking method for big dynamic crowds or numerous pedestrians. Three modules constitute the proposed method, i.e., face candidate generation, face candidate verification, and face target tracking. In this work, face candidates are localized using the features of the face area, edge information, and skin color. Non-face parts in the face candidates are further verified by the C-SVM learning model and then removed, by which the face targets can be generated with lower computation-complexity and satisfactory accuracy than other approaches. Finally, the face targets are tracked by an efficient and reliable searching scheme for improving the effective face detection rate. Experimental results show that the average face detection rate (FDR) of 85%, average effective FDR of 95%, a frame rate of 28–66 frames per second (fps), and about 30 faces detected per frame are obtained from various test videos with big dynamic crowds or numerous pedestrians, indicating the feasibility of the proposed method to achieve unconstrained face detection with highefficiency and cost-effectiveness. This result makes the proposed method more attractive for the video surveillance system as compared to other approaches, especially in the high computational complexity-based methods. Keywords Face detection . Face tracking . Numerous pedestrians . Histogram of oriented gradient (HOG) . Support vector machine (SVM)
1 Introduction Face detection is a very hot topic in the research fields of computer vision and also widely applied to image-based security and surveillance systems, identity authentication, crime * Chao-Ho Chen [email protected] Extended author information available on the last page of the article
Multimedia Tools and Applications
prevention, pedestrian tracking, and human-machine interaction so on. Generally, face detection mainly involves finding a face position (i.e., face localization) and recognizing face candidate to be a real face or not (i.e., face determination). Thus far, face detection in big dynamic crowds or numerous pedestrians still remains a great challenging task due to variations in locations, scales, postures (i.e., different face view angles), expressions, crowd density, and varying illuminations, under the consideration of cost-effectiveness. By using a fixed surveillance camera, faces of varying sizes and view angles can be captured depending on the distance from the camera. That is, the closer the camera is, the larger the face size will be, and vice versa. For a dynamic crowd of people in public space or numerous pedestrians, many faces and smaller face sizes greatly
Data Loading...
