People detection in surveillance: classification and evaluation

Abstract

Nowadays people detection in video surveillance environments is a task that has been generating great interest. There are many approaches trying to solve the problem either in controlled scenarios or in very specific surveillance applications. Themain objective of this paper is to give a comprehensive and extensive evaluation of the state of the art of people detection regardless of the final surveillance application. For this reason, firstly, the different processing tasks involved in the automatic people detection in video sequences have been defined, then a proper classification of the state of the art of people detection has been made according to the two most critical tasks, object detection and person model, that are needed in every detection approach. Finally, experiments have been performed on an extensive dataset with different approaches that completely cover the proposed classification and support the conclusions drawn from the state of the art. Introduction:Computer vision has been an evolving field for the last years with multiple lines of research and different application domains. Video surveillance has been one of the most developed domains for the last 10 years [1,2,3,4]. The need for providing security to people and their properties in the entire world explains the huge development and expansion of video surveillance systems nowadays. Video surveillance systems try to automatically extract information from the video sequence and to generate a scene description useful for human interactions with the system: alarms, logs, statistics, indexing and retrieval, etc.Within the computer vision field, particularly in the research area of digital image and video processing, there exists a rich variety of algorithms for segmentation, object detection, event recognition, etc, which are being used in surveillance systems. Automatic people detection in video sequences [5,6,7,8] is one of the most challenging problems in computer vision. The complexity of the people detection problem is mainly based on the difficulty of modeling people because of their huge variability in physical appearances, articulated body parts, poses, movements, points of view and interactions among different people and objects. This complexity is even higher in typical real world surveillance scenarios such as airports, malls, etc, which often include multiple people, multiple occlusions and background variability.There is a large number of people detection surveys in the literature, some of them partially cover only the state of the art or are clearly focused on some particular video surveillance application.[5] presents a survey of people detection and also the integration of the detectors into onboard full systems. It decomposes people detection approaches into three processing tasks: generation of initial object hypotheses or Regions of Interest (ROI) selection, verification (classification) and temporal integration (tracking).[6] also presents a survey of people detection but with a clear focus on driver assistance systems and def...

Multiple classifiers in biometrics. Part 2: Trends and challenges

“…The presentation there was general with a generic mathematical formulation, in order to facilitate the export of experiences and methods to other information fusion problems, e.g. : video surveillance [3], speech technologies [4], biomedical applications [91], human-computer interaction [5], data analytics [6], or recommender systems [7]. Part 1 was intended for the non-expert in MCS, or any other reader interested in overviewing the field of multimodal biometrics.…”

“…Most traditional and widely deployed biometric solutions for person recognition are designed for access control or forensic scenarios. One important challenge in biometrics is how to properly integrate biometrics technologies in other application scenarios like mobile authentication [71,72], video surveillance [3], forensics [73], largescale ID [74], cloud biometrics or ubiquitous biometrics [75].…”

“…Part 1 then overviewed the application of MCS to the particular field of multimodal biometric person authentication in the last 25 years [2], as a prototypical area in which MCS has resulted in important achievements. That overview of MCS applied to biometrics was developed using a generic MCS framework and mathematical notation, with the purpose of facilitating the transfer of MCS achievements from here to other pattern recognition applications like video surveillance [3], speech technologies [4], human-computer interaction [5], data analytics applications [6], or recommender systems [7].…”

Real Time Advanced Clustering System

“…A clustering system, in particular regarding detection and tracking of moving objects, is a really common and important task, widely used in computer vision as low-level task for different applications like: camera surveillance [1], Cross Traffic Alert (CTA) [2], Intelligent Transport System (ITS) [3], object detection and recognition [4], people detection and recognition [5], people behavior analysis (e. g. anti-theft, falling detection, etc.) [6] and user interfaces by gesture [7].…”

Feature Encodings and Poolings for Action and Event Recognition: A Comprehensive Survey

“…There are many promising applications for action and event recognition, such as abnormal action and event recognition in surveillance applications [1][2][3][4], interaction action and event recognition in entertainment applications [5][6][7][8], and home-based rehabilitation action and event recognition in healthcare applications [9][10][11][12], and many other analogous applications such as in [13][14][15][16][17][18]. According to the definition given by NIST [19], an event is a complex activity occurring at a specific place and time, which involves people interacting with other people and/or objects, and consists of a number of human actions, processes, and activities.…”