Showing papers on "Activity recognition published in 2002"

Layered representations for human activity recognition

Abstract: We present the use of layered probabilistic representations using Hidden Markov Models for performing sensing, learning, and inference at multiple levels of temporal granularity. We describe the use of the representation in a system that diagnoses states of a user's activity based on real-time streams of evidence from video, acoustic, and computer interactions. We review the representation, present an implementation, and report on experiments with the layered representation in an office-awareness application.

Human activity recognition using multidimensional indexing

Abstract: In this paper, we develop a novel method for view-based recognition of human action/activity from videos. By observing just a few frames, we can identify the activity that takes place in a video sequence. The basic idea of our method is that activities can be positively identified from a sparsely sampled sequence of a few body poses acquired from videos. In our approach, an activity is represented by a set of pose and velocity vectors for the major body parts (hands, legs, and torso) and stored in a set of multidimensional hash tables. We develop a theoretical foundation that shows that robust recognition of a sequence of body pose vectors can be achieved by a method of indexing and sequencing and it requires only a few pose vectors (i.e., sampled body poses in video frames). We find that the probability of false alarm drops exponentially with the increased number of sampled body poses. So, matching only a few body poses guarantees high probability for correct recognition. Our approach is parallel, i.e., all possible model activities are examined at one indexing operation. In addition, our method is robust to partial occlusion since each body part is indexed separately. We use a sequence-based voting approach to recognize the activity invariant to the activity speed.

Recognizing and Tracking Human Action

Abstract: Human activity can be described as a sequence of 3D body postures. The traditional approach to recognition and 3D reconstruction of human activity has been to track motion in 3D, mainly using advanced geometric and dynamic models. In this paper we reverse this process. View based activity recognition serves as an input to a human body location tracker with the ultimate goal of 3D reanimation in mind. We demonstrate that specific human actions can be detected from single frame postures in a video sequence. By recognizing the image of a person's posture as corresponding to a particular key frame from a set of stored key frames, it is possible to map body locations from the key frames to actual frames. This is achieved using a shape matching algorithm based on qualitative similarity that computes point to point correspondence between shapes, together with information about appearance. As the mapping is from fixed key frames, our tracking does not suffer from the problem of having to reinitialise when it gets lost. It is effectively a closed loop. We present experimental results both for recognition and tracking for a sequence of a tennis player.

Wearable sensing to annotate meeting recordings

Abstract: We propose to use wearable computers and sensor systems to generate personal contextual annotations in audio visual recordings of meetings. In this paper we argue that such annotations are essential and effective to allow retrieval of relevant information from large audio-visual databases. The paper proposes several useful annotations that can be derived from cheap and unobtrusive sensors. It also describes a hardware platform designed to implement this concept and presents first experimental results.

Automated Visual Surveillance Using Hidden Markov Models

Abstract: This paper describes an automated visual surveillance system that detects suspicious human activity in a scene. The system is designed to: 1) detect and track people in the scene, 2) recognize the “normal” activities in the scene, and 3) detect anomalous activity by finding sufficiently large deviations from the normal activity patterns. The stochastic time-sequence recognition framework of the Hidden Markov Model (HMM) forms the basis of activity recognition and anomaly detection. We have implemented the system to monitor an office corridor in real-time using a Pentium III machine running Windows 2000. The results show that the system correctly classifies examples of normal activities in the corridor and identifies a mock break-in attempt as suspicious activity.

A Survey of Visual Analysis of Human Motion

Abstract: Visual analysis of human motion is currently one of the most active research topics in the domain of computer vision. This strong interest is driven by a wide spectrum of promising applications in many areas such as virtual reality, smart surveillance, perceptual user interface, content based image storage and retrieval, athletic performance analysis, etc. Human motion analysis aims at attempting to detect, track and identify people, and more generally, to understand human behaviors, from image sequences involving humans. This paper provides a comprehensive survey of recent developments of vision based human motion analysis, and it keeps up with the latest research ranging mainly from 1995 to 2000. Different from previous reviews, our emphasis is on four major issues involved in a general framework of human motion analysis, namely motion detection, moving object classification, human tracking, and activity recognition and description. This paper focuses more on overall methods and general characteristics involved in the above four issues, so each issue is accordingly divided into sub processes and categories of approaches so as to provide more detailed discussions. At first, we introduce some potential applications of human motion analysis. Then, various existing methods for each key issue are clearly discussed to examine the state of the art in human motion analysis. Motion detection provides a focus of attention for later processes because only those changing pixels need be considered. Three types of techniques are addressed, namely background subtraction, temporal differencing and optical flow. As far as moving object classification is concerned, shape based or motion based methods are presented. Tracking is equivalent to establishing correspondence of image features between consecutive frames, and four approaches studied intensively in past work are described:model based, active contour based, region based and feature based. The task of recognizing human activity in image sequences assumes that feature tracking for recognition has been accomplished. Two types of techniques, template matching and state space approaches, are reviewed. Although a large amount of work has been done in the field of human motion analysis, many issues still remain open such as segmentation, modeling and occlusion handling. At the end of this survey, some detailed discussions on research challenges and future directions in human motion analysis are also provided. Past achievements show to some extent that vision systems have considerable ability to cope with complex human movements, so we are looking forward to more new techniques to devote this field.

Probabilistic motion parameter models for human activity recognition

Abstract: A novel method for human activity recognition is presented. Given a video sequence containing human activity, the motion parameters of each frame are first computed using different motion parameter models. The likelihood of these observed motion parameters is optimally approximated, based directly on a multivariate Gaussian probabilistic model. The dynamic change of motion parameter likelihood in a video sequence is characterized using a continuous density hidden Markov model. Activity recognition is then posed as a motion parameter maximum likelihood estimation problem. Experimental results show that the method proposed here works well in recognizing such complex human activities as sitting, getting up from a chair, and some martial art actions.

A hierarchical human detection system in (un)compressed domains

Abstract: We propose a hierarchical retrieval system where shape, color and motion characteristics of the human body are captured in compressed and uncompressed domains. The proposed retrieval method provides human detection and activity recognition at different resolution levels from low complexity to low false rates and connects low level features to high level semantics by developing relational object and activity presentations. The available information of standard video compression algorithms are used in order to reduce the amount of time and storage needed for the information retrieval. The principal component analysis is used for activity recognition using MPEG motion vectors and results are presented for walking, kicking, and running to demonstrate that the classification among activities is clearly visible. For low resolution and monochrome images it is demonstrated that the structural information of human silhouettes can be captured from AC-DCT coefficients.

Continuous activity recognition with missing data

Abstract: Human activity recognition involves several problems like changes when an activity is performed by different persons. This means that people can perform the same activity faster or slower and also the way that an activity is performed can change, therefore we can have different trajectories representing the same activity. Another problem exists when we do not have the whole trajectory because of occlusion or noise. In this work, an approach for human activity recognition based on the Fourier transform and Bayesian networks is presented. This approach can recognize activities performed at different velocities by different people and can work with missing data. It performs continuous activity recognition without the necessity of manually indicating when the activity starts or finishes.

A bottom-up approach for activity recognition in smart rooms

Abstract: We propose a smart camera system where the cameras detect the presence of a person and recognize activities of this person. A relational graph-based modeling of human body and a HMM-based activity recognition of the body parts are proposed for real-time video analysis. The results show that more than 86 percent of the body parts and 88 percent of the activities are correctly classified. We also describe the relationship between the activity detection algorithms and the architectures required to perform these tasks in real time. We achieve a processing rate of more than 20 frames per second for each TriMedia video capture board.

Context-aware activity recognition using TAN classifiers

Abstract: This thesis reviews the components necessary for designing and implementing a realtime activity recognition system for mobile computing devices. In particular, a system utilizing GPS location data and tree augmented naive Bayes (TAN) classifiers is described and evaluated. The system can successfully recognize activities such as shopping, going to work, returning home, and going to a restaurant. Several different sets of features are tested using both the TAN algorithm and a test bed of other competitive classifiers. Experimental results show that the system can recognize about 85% of activities correctly using a multinet version of the TAN algorithm. Although efforts were made to design a general-purpose system, findings indicate that the nature of the position data and many relevant features are person-specific. The results from this research provide a foundation upon which future activity aware applications can be built. Thesis Supervisor: Stephen S. Intille Title: Research Scientist