Probability Distributions.- Linear Models for Regression.- Linear Models for Classification.- Neural Networks.- Kernel Methods.- Sparse Kernel Machines.- Graphical Models.- Mixture Models and EM.- Approximate Inference.- Sampling Methods.- Continuous Latent Variables.- Sequential Data.- Combining Models.

Pattern Recognition and Machine Learning

The particle filter (PF) was introduced in 1993 as a numerical approximation to the nonlinear Bayesian filtering problem, and there is today a rather mature theory as well as a number of successful applications described in literature. This tutorial serves two purposes: to survey the part of the theory that is most important for applications and to survey a number of illustrative positioning applications from which conclusions relevant for the theory can be drawn. The theory part first surveys the nonlinear filtering problem and then describes the general PF algorithm in relation to classical solutions based on the extended Kalman filter (EKF) and the point mass filter (PMF). Tuning options, design alternatives, and user guidelines are described, and potential computational bottlenecks are identified and remedies suggested. Finally, the marginalized (or Rao-Blackwellized) PF is overviewed as a general framework for applying the PF to complex systems. The application part is more or less a stand-alone tutorial without equations that does not require any background knowledge in statistics or nonlinear filtering. It describes a number of related positioning applications where geographical information systems provide a nonlinear measurement and where it should be obvious that classical approaches based on Kalman filters (KFs) would have poor performance. All applications are based on real data and several of them come from real-time implementations. This part also provides complete code examples.

/pdf/particle-filter-theory-and-practice-with-positioning-3ub8asx2jb.pdf

Particle filter theory and practice with positioning applications

Linköping University Post Print Particle Filter Theory and Practice with Positioning Applications

Low-rank matrices play a fundamental role in modeling and computational methods for signal processing and machine learning. In many applications where low-rank matrices arise, these matrices cannot be fully sampled or directly observed, and one encounters the problem of recovering the matrix given only incomplete and indirect observations. This paper provides an overview of modern techniques for exploiting low-rank structure to perform matrix recovery in these settings, providing a survey of recent advances in this rapidly-developing field. Specific attention is paid to the algorithms most commonly used in practice, the existing theoretical guarantees for these algorithms, and representative practical applications of these techniques.

An Overview of Low-Rank Matrix Recovery From Incomplete Observations

Motion blur due to camera shake is one of the predominant sources of degradation in handheld photography. Single image blind deconvolution (BD) or motion deblurring aims at restoring a sharp latent image from the blurred recorded picture without knowing the camera motion that took place during the exposure. BD is a long-standing problem, but has attracted much attention recently, cumulating in several algorithms able to restore photos degraded by real camera motion in high quality. In this paper, we present a benchmark dataset for motion deblurring that allows quantitative performance evaluation and comparison of recent approaches featuring non-uniform blur models. To this end, we record and analyse real camera motion, which is played back on a robot platform such that we can record a sequence of sharp images sampling the six dimensional camera motion trajectory. The goal of deblurring is to recover one of these sharp images, and our dataset contains all information to assess how closely various algorithms approximate that goal. In a comprehensive comparison, we evaluate state-of-the-art single image BD algorithms incorporating uniform and non-uniform blur models.

/pdf/recording-and-playback-of-camera-shake-benchmarking-blind-2gpc6mbydk.pdf

Recording and playback of camera shake: benchmarking blind deconvolution with a real-world database

Blind Image Deconvolution

A classical approach to abnormal activity detection is to learn a representation for normal activities from the training data and then use this learned representation to detect abnormal activities while testing. Typically, the methods based on this approach operate at a fixed timescale — either a single time-instant (e.g. frame-based) or a constant time duration (e.g. video-clip based). But human abnormal activities can take place at different timescales. For example, jumping is a short-term anomaly and loitering is a long-term anomaly in a surveillance scenario. A single and pre-defined timescale is not enough to capture the wide range of anomalies occurring with different time duration. In this paper, we propose a multi-timescale model to capture the temporal dynamics at different timescales. In particular, the proposed model makes future and past predictions at different timescales for a given input pose trajectory. The model is multi-layered where intermediate layers are responsible to generate predictions corresponding to different timescales. These predictions are combined to detect abnormal activities. In addition, we also introduce a single-camera abnormal activity dataset for research use that contains 483,566 annotated frames. Our experiments show that the proposed model can capture the anomalies of different time duration and outperforms existing methods.

/pdf/multi-timescale-trajectory-prediction-for-abnormal-human-4b8qlq91n9.pdf

Multi-timescale Trajectory Prediction for Abnormal Human Activity Detection

In this paper, we propose a particle filter acoustic direction-of-arrival (DOA) tracker to track multiple maneuvering targets using a state space approach. The particle filter determines its state vector using a batch of DOA estimates. The filter likelihood treats the observations as an image, using template models derived from the state update equation, and also incorporates the possibility of missing data as well as spurious DOA observations. Multiple targets are handled using a partitioned state-vector approach. The particle filter solution is compared with three other methods: the extended Kalman filter, Laplacian filter, and another particle filter that uses the acoustic microphone outputs directly. In addition, we demonstrate an autonomous system for multiple target DOA tracking with automatic target initialization and deletion. The initialization system uses a track-before-detect approach and employs matching pursuit to initialize multiple targets. Computer simulations are presented to compare the performance of the algorithms

/pdf/acoustic-multitarget-tracking-using-direction-of-arrival-2ifqfn5bby.pdf

Acoustic Multitarget Tracking Using Direction-of-Arrival Batches

Researchers have been working on the blind deconvolution problem from as early as 1975 and various methods have been developed over the period. Earlier methods were purely transform domain based, which were modifications of the inverse filter. This was followed by solutions that treat blind deconvolution as an ill-posed problem, thereby using regularization as a tool for solving the problem. Various refinements to this approach that have been proposed are reviewed in this chapter. We also devote a separate section to motion deblur, since there is a renewed interest in this area and a large volume of publications has come up recently in this area.

Blind Deconvolution Methods: A Review

We propose a particle filter tracker to track multiple maneuvering targets using a batch of range measurements. The state update is formulated through a locally linear motion model and the observability of the state vector is proved using geometrical arguments. The data likelihood treats the range observations as an image using template models derived from the state update equation, and incorporates the possibility of missing data as well as spurious range observations. The particle filter handles multiple targets, using a partitioned state-vector approach. The filter proposal function uses a Gaussian approximation of the full-posterior to cope with target maneuvers for improved efficiency. By treating the range measurements as images and using smoothness constraints, the particle filter is able to avoid the data association problems. Computer simulations demonstrate the performance of the tracking algorithm.

/pdf/a-range-only-multiple-target-particle-filter-tracker-1yvr6jjb80.pdf

Rajbabu Velmurugan

Papers

Blind Image Deconvolution

Multi-timescale Trajectory Prediction for Abnormal Human Activity Detection

Acoustic Multitarget Tracking Using Direction-of-Arrival Batches

Blind Deconvolution Methods: A Review

A Range-Only Multiple Target Particle Filter Tracker