On the sustained tracking of human motion

2000

We propose a novel hierarchical model of human dynamics for view independent tracking of the human body in monocular video sequences. The model is trained using real data from a collection of people. Kinematics are encoded using Hierarchical Principal Component Analysis, and dynamics are encoded using Hidden Markov Models. The top of the hierarchy contains information about the whole body. The lower levels of the hierarchy contain more detailed information about possible poses of some subpart of the body. When tracking, the lower levels of the hierarchy are shown to improve accuracy. In this article we describe our model and present experiments that show we can recover 3D skeletons from 2D images in a view independent manner, and also track people the system was not trained on.

2004

We present a method to solve the human silhouette tracking problem using 18 major human points. We used: a simple 2D model for the human silhouette, a linear prediction technique for initializing major points search, geometry anthropometric constraints for determining the search area and color measures for matching human body parts. In addition, we propose a method to solve the problem of human members recognition and 18 major human points detection using silhouette. This result can be used to initialize a human tracking algorithm for real time applications. Our main purpose is to develop a low computation cost algorithm, which can be used independently of camera motion. The output of the tracking algorithm is the position of 18 major human points and a 2D human body extraction. In cases of low quality imaging conditions or low background contrast, the result may be worst. For these cases we defined an appropriate criterion concerning tracking ability.

IASET, 2013

Video surveillances system based on human detection, tracking and human pose estimation assurances to be an important technology for real time applications, including the analysis of human activities. So many applications have been demonstrated regarding this technology but evaluations of some key features are remains challenging. How human being going to be detected that is the challenge i.e. According to body structure, skin color, skeleton, etc. It is very complex because each human being has different human kinematic structure, variation in body size and shape. In the human tracking process occlusions of body parts, inabilities to observe the skeletal motion due to clothing, difficulty segmenting the human from the background these are the challenges. Pose estimation is also challenging because complex interactions between people in the environment, clothing complicates the skeleton structure, and significantly increases the inconsistency of individual human appearance. Some image related components also increases the challenges because limited image resolution, number of ambiguities, and the inability to easily distinguish the parts of a human from occlusion or from the cluttered background. With the prior knowledge some of these challenges are resolved, but some of the problems require clever mathematical and engineering solutions. KEYWORDS: Discriminative Methods for Pose Estimation, Human

Computer Vision and Image Understanding, 2001

Tracking and modeling people from video sequences has become an increasingly important research topic, with applications including animation, surveillance and sports medicine. In this paper, we propose a model based 3-D approach to recovering both body shape and motion. It takes advantage of a sophisticated animation model to achieve both robustness and realism. Stereo sequences of people in motion serve as input to our system. From these, we extract a 2 1 2 -D description of the scene and, optionally, silhouette edges. We propose an integrated framework to fit the model and to track the person's motion. The environment does not have to be engineered. We recover not only the motion but also a full animation model closely resembling the subject. We present results of our system on real sequences and we show the generic model adjusting to the person and following various kinds of motion.

International Conference on Pattern Recognition, 2002

A human appearance modelling framework where colour distributions are associated with surface regions on an ar- ticulated body model is presented. In general, these dis- tributions are unknown, multi-modal and changing in time. We therefore propose using recursively updated histograms to represent them. For a certain pose, a set of histograms may be collected and a likelihood constructed based on

Humanoid Robots: New Developments, 2007

2016

An extensive amount of research is undergoing in the field of computer vision is related to object detection & tracking. Detection and tracking of the moving objects in dynamic environment is a difficult task. In this paper, we describe a framework for the detection and tracking of moving people in indoor\outdoor environment. The algorithm consists of two steps: at first, we use a Dual frame differencing method for motion detection. Then, algorithm performance is improved for false motion detection using morphological operations. The second step of the algorithm is object tracking framework based on Kalman filtering which works well in dynamic scenes. Experimental result shows that the method yields superior performance as compared to the other methods of video surveillance. Key-words: Video surveillance; Object Tracking; Kalman Filter; Dual Frame Differences.

2005

Tracking and recognition of human motion has become an important research area in computer vision. In real-world conditions it constitutes a complicated problem, considering cluttered backgrounds, gross illumination variations, occlusions, self-occlusions, different clothing, and multiple moving objects. These ill-posed problems are usually tackled by simplifying assumptions regarding the scene or by imposing constraints on the motion. Constraints such as that the contrast between the moving people and the background ...

Computer Vision and Image Understanding, 2006

1995

INTRODUCTIONComputer-assisted surveillance of complexenvironments is getting more and more interesting,thanks to recent significant improvements in real-timesignal processing. The main role of automaticcomputation in such systems is to support the humanoperator to perform many tasks, such as detecting,interpreting and understanding, logging or givingalarms. In this way, possible human failures areexpected to be overcome, many functionalities can beadded, and better surveillance...