Results 1  10
of
166
Recognition of group activities using dynamic probabilistic networks
 Computer Vision, 2003. Proceedings. Ninth IEEE International Conference
"... Dynamic Probabilistic Networks (DPNs) are exploited for modelling the temporal relationships among a set of different object temporal events in the scene for a coherent and robust scenelevel behaviour interpretation. In particular, we develop a Dynamically MultiLinked Hidden Markov Model (DMLHM ..."
Abstract

Cited by 127 (22 self)
 Add to MetaCart
(Show Context)
Dynamic Probabilistic Networks (DPNs) are exploited for modelling the temporal relationships among a set of different object temporal events in the scene for a coherent and robust scenelevel behaviour interpretation. In particular, we develop a Dynamically MultiLinked Hidden Markov Model (DMLHMM) to interpret group activities involving multiple objects captured in an outdoor scene. The model is based on the discovery of salient dynamic interlinks among multiple temporal events using DPNs. Object temporal events are detected and labelled using Gaussian Mixture Models with automatic model order selection. A DMLHMM is built using Schwarz’s Bayesian Information Criterion based factorisation resulting in its topology being intrinsically determined by the underlying causality and temporal order among different object events. Our experiments demonstrate that its performance on modelling group activities in a noisy outdoor scene is superior compared to that of a MultiObservation Hidden Markov Model (MOHMM), a Parallel Hidden Markov Model (PaHMM) and a Coupled
Inferring Networks of Diffusion and Influence
, 2010
"... Information diffusion and virus propagation are fundamental processes talking place in networks. While it is often possible to directly observe when nodes become infected, observing individual transmissions (i.e., who infects whom or who influences whom) is typically very difficult. Furthermore, in ..."
Abstract

Cited by 116 (13 self)
 Add to MetaCart
Information diffusion and virus propagation are fundamental processes talking place in networks. While it is often possible to directly observe when nodes become infected, observing individual transmissions (i.e., who infects whom or who influences whom) is typically very difficult. Furthermore, in many applications, the underlying network over which the diffusions and propagations spread is actually unobserved. We tackle these challenges by developing a method for tracing paths of diffusion and influence through networks and inferring the networks over which contagions propagate. Given the times when nodes adopt pieces of information or become infected, we identify the optimal network that best explains the observed infection times. Since the optimization problem is NPhard to solve exactly, we develop an efficient approximation algorithm that scales to large datasets and in practice gives provably nearoptimal performance. We demonstrate the effectiveness of our approach by tracing information cascades in a set of 170 million blogs and news articles over a one year period to infer how information flows through the online media space. We find that the diffusion network of news tends to have a coreperiphery structure with a small set of core media sites that diffuse information to the rest of the Web. These sites tend to have stable circles of influence with more general news media sites acting as connectors between them.
Beyond tracking: modelling activity and understanding behaviour
 International Journal of Computer Vision
, 2006
"... In this work, we present a unified bottomup and topdown automatic model selection based approach for modelling complex activities of multiple objects in cluttered scenes. An activity of multiple objects is represented based on discrete scene events and their behaviours are modelled by reasoning ab ..."
Abstract

Cited by 79 (14 self)
 Add to MetaCart
(Show Context)
In this work, we present a unified bottomup and topdown automatic model selection based approach for modelling complex activities of multiple objects in cluttered scenes. An activity of multiple objects is represented based on discrete scene events and their behaviours are modelled by reasoning about the temporal and causal correlations among different events. This is significantly different from the majority of the existing techniques that are centred on object tracking followed by trajectory matching. In our approach, objectindependent events are detected and classified by unsupervised clustering using ExpectationMaximisation (EM) and classified using automatic model selection based on Schwarz’s Bayesian Information Criterion (BIC). Dynamic Probabilistic Networks (DPNs) are formulated for modelling the temporal and causal correlations among discrete events for robust and holistic scenelevel behaviour interpretation. In particular, we developed a Dynamically MultiLinked Hidden Markov Model (DMLHMM) based on the discovery of salient dynamic interlinks among multiple temporal processes corresponding to multiple event classes. A DMLHMM is built using BIC based factorisation resulting in its topology being intrinsically determined by the underlying causality and temporal order among events. Extensive experiments are conducted on modelling activities captured in different indoor and
Online filtering, smoothing and probabilistic modeling of streaming data
 in ICDE
, 2008
"... In this paper, we address the problem of extending a relational database system to facilitate efficient realtime application of dynamic probabilistic models to streaming data. We use the recently proposed abstraction of modelbased views for this purpose, by allowing users to declaratively specify ..."
Abstract

Cited by 69 (3 self)
 Add to MetaCart
(Show Context)
In this paper, we address the problem of extending a relational database system to facilitate efficient realtime application of dynamic probabilistic models to streaming data. We use the recently proposed abstraction of modelbased views for this purpose, by allowing users to declaratively specify the model to be applied, and by presenting the output of the models to the user as a probabilistic database view. We support declarative querying over such views using an extended version of SQL that allows for querying probabilistic data. Underneath we use particle filters, a class of sequential Monte Carlo algorithms commonly used to implement dynamic probabilistic models, to represent the present and historical states of the model as sets of weighted samples (particles) that are kept uptodate as new readings arrive. We develop novel techniques to convert the queries on the modelbased view directly into queries over particle tables, enabling highly efficient query processing. Finally, we present experimental evaluation of our prototype implementation over sensor data from the Intel Lab dataset that demonstrates the feasibility of online modeling of streaming data using our system and establishes the advantages of such tight integration between dynamic probabilistic models and database systems. 1
Switching Kalman Filters
, 1998
"... We show how many different variants of Switching Kalman Filter models can be represented in a unified way, leading to a single, generalpurpose inference algorithm. We then show how to find approximate Maximum Likelihood Estimates of the parameters using the EM algorithm, extending previous results ..."
Abstract

Cited by 67 (2 self)
 Add to MetaCart
We show how many different variants of Switching Kalman Filter models can be represented in a unified way, leading to a single, generalpurpose inference algorithm. We then show how to find approximate Maximum Likelihood Estimates of the parameters using the EM algorithm, extending previous results on learning using EM in the nonswitching case [DRO93, GH96a] and in the switching, but fully observed, case [Ham90]. 1 Introduction Dynamical systems are often assumed to be linear and subject to Gaussian noise. This model, called the Linear Dynamical System (LDS) model, can be defined as x t = A t x t\Gamma1 + v t y t = C t x t +w t where x t is the hidden state variable at time t, y t is the observation at time t, and v t ¸ N(0; Q t ) and w t ¸ N(0; R t ) are independent Gaussian noise sources. Typically the parameters of the model \Theta = f(A t ; C t ; Q t ; R t )g are assumed to be timeinvariant, so that they can be estimated from data using e.g., EM [GH96a]. One of the main adva...
MEBN: A Language for FirstOrder Bayesian Knowledge Bases
"... Although classical firstorder logic is the de facto standard logical foundation for artificial intelligence, the lack of a builtin, semantically grounded capability for reasoning under uncertainty renders it inadequate for many important classes of problems. Probability is the bestunderstood and m ..."
Abstract

Cited by 65 (24 self)
 Add to MetaCart
(Show Context)
Although classical firstorder logic is the de facto standard logical foundation for artificial intelligence, the lack of a builtin, semantically grounded capability for reasoning under uncertainty renders it inadequate for many important classes of problems. Probability is the bestunderstood and most widely applied formalism for computational scientific reasoning under uncertainty. Increasingly expressive languages are emerging for which the fundamental logical basis is probability. This paper presents MultiEntity Bayesian Networks (MEBN), a firstorder language for specifying probabilistic knowledge bases as parameterized fragments of Bayesian networks. MEBN fragments (MFrags) can be instantiated and combined to form arbitrarily complex graphical probability models. An MFrag represents probabilistic relationships among a conceptually meaningful group of uncertain hypotheses. Thus, MEBN facilitates representation of knowledge at a natural level of granularity. The semantics of MEBN assigns a probability distribution over interpretations of an associated classical firstorder theory on a finite or countably infinite domain. Bayesian inference provides both a proof theory for combining prior knowledge with observations, and a learning theory for refining a representation as evidence accrues. A proof is given that MEBN can represent a probability distribution on interpretations of any finitely axiomatizable firstorder theory.
Finding Your Friends and Following Them to Where You Are
"... Location plays an essential role in our lives, bridging our online and offline worlds. This paper explores the interplay between people’s location, interactions, and their social ties within a large realworld dataset. We present and evaluate Flap, a system that solves two intimately related tasks: ..."
Abstract

Cited by 65 (11 self)
 Add to MetaCart
(Show Context)
Location plays an essential role in our lives, bridging our online and offline worlds. This paper explores the interplay between people’s location, interactions, and their social ties within a large realworld dataset. We present and evaluate Flap, a system that solves two intimately related tasks: link and location prediction in online social networks. For link prediction, Flap infers social ties by considering patterns in friendship formation, the content of people’s messages, and user location. We show that while each component is a weak predictor of friendship alone, combining them results in a strong model, accurately identifying the majority of friendships. For location prediction, Flap implements a scalable probabilistic model of human mobility, where we treat users with known GPS positions as noisy sensors of the location of their friends. We explore supervised and unsupervised learning scenarios, and focus on the efficiency of both learning and inference. We evaluate Flap on a large sample of highly active users from two distinct geographical areas and show that it (1) reconstructs the entire friendship graph with high accuracy even when no edges are given; and (2) infers people’s finegrained location, even when they keep their data private and we can only access the location of their friends. Our models significantly outperform current comparable approaches to either task.
Video behaviour profiling and abnormality detection without manual labelling
 In IEEE International Conference on Computer Vision
, 2005
"... A novel framework is developed for automatic behaviour profiling and abnormality sampling/detection without any manual labelling of the training dataset. Natural grouping of behaviour patterns is discovered through unsupervised model selection and feature selection on the eigenvectors of a normalise ..."
Abstract

Cited by 57 (6 self)
 Add to MetaCart
(Show Context)
A novel framework is developed for automatic behaviour profiling and abnormality sampling/detection without any manual labelling of the training dataset. Natural grouping of behaviour patterns is discovered through unsupervised model selection and feature selection on the eigenvectors of a normalised affinity matrix. Our experiments demonstrate that a behaviour model trained using an unlabelled dataset is superior to those trained using the same but labelled dataset in detecting abnormality from an unseen video. 1.
On the Convexity of Latent Social Network Inference
"... In many realworld scenarios, it is nearly impossible to collect explicit social network data. In such cases, whole networks must be inferred from underlying observations. Here, we formulate the problem of inferring latent social networks based on network diffusion or disease propagation data. We co ..."
Abstract

Cited by 55 (4 self)
 Add to MetaCart
(Show Context)
In many realworld scenarios, it is nearly impossible to collect explicit social network data. In such cases, whole networks must be inferred from underlying observations. Here, we formulate the problem of inferring latent social networks based on network diffusion or disease propagation data. We consider contagions propagating over the edges of an unobserved social network, where we only observe the times when nodes became infected, but not who infected them. Given such node infection times, we then identify the optimal network that best explains the observed data. We present a maximum likelihood approach based on convex programming with a l1like penalty term that encourages sparsity. Experiments on real and synthetic data reveal that our method nearperfectly recovers the underlying network structure as well as the parameters of the contagion propagation model. Moreover, our approach scales well as it can infer optimal networks on thousand nodes in a matter of minutes. 1
Bayesian Modality Fusion: Probabilistic Integration Of Multiple Vision Algorithms for Head Tracking
 FOURTH ASIAN CONFERENCE ON COMPUTER VISION (ACCV)
, 2000
"... We describe a headtracking system that harnesses Bayesian modality fusion, a technique for integrating the analyses of multiple visual tracking algorithms within a probabilistic framework. At the heart of the approach is a Bayesian network model that includes random variables that serve as context ..."
Abstract

Cited by 55 (4 self)
 Add to MetaCart
We describe a headtracking system that harnesses Bayesian modality fusion, a technique for integrating the analyses of multiple visual tracking algorithms within a probabilistic framework. At the heart of the approach is a Bayesian network model that includes random variables that serve as contextsensitive indicators of reliability of the different tracking algorithms. Parameters of the Bayesian model are learned from data in an offline training phase using groundtruth data from a Polhemus tracking device. In our implementation for a realtime head tracking task, algorithms centering on color, motion, and background subtraction modalities are fused into a single estimate of head position in an image. Results demonstrate the effectiveness of Bayesian modality fusion in environments undergoing a variety of visual perturbances.