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 PASCAL Visual Object Classes Challenge

In fluorescence correlation spectroscopy (FCS), information about molecular dynamics is extracted by recording the fluctuating signal of a pico- to micromolar concentration of fluorescent molecules in solution. As primary parameters, FCS provides time constants of the fluorescence emission, as well as numbers and dwell times of the observed particles diffusing through the open volume. A biochemical reaction or macromolecular rearrangement causing changes in these parameters, when linked to fluorescence readout, can be quantified by FCS. Since the measurements are now routinely performed in a laser-illuminated confocal spot, making measurements in living cells is straightforward. Different cellular compartments, such as the nucleus, the cytoplasm, or the plasma membrane, can be easily distinguished and addressed. In addition to biochemical reactions, the local environment of macromolecules, for example, ion concentrations, pH, or viscosity, can be probed. Thus, FCS is a versatile and attractive technique for researchers striving for a quantitative understanding of interactions and dynamics of biological and in particular cellular systems.

Fluorescence Correlation Spectroscopy in Living Cells

The motion history image (MHI) approach is a view-based temporal template method which is simple but robust in representing movements and is widely employed by various research groups for action recognition, motion analysis and other related applications. In this paper, we provide an overview of MHI-based human motion recognition techniques and applications. Since the inception of the MHI template for motion representation, various approaches have been adopted to improve this basic MHI technique. We present all important variants of the MHI method. This paper points some areas for further research based on the MHI method and its variants.

Motion history image: its variants and applications

Methods to measure the lateral diffusion of membrane lipids and proteins.

Molecular dynamics of STAT3 on IL-6 signaling pathway in living cells.

Logistic label propagation

Extracting in-play scenes in sport videos is essential for quantitative analysis and effective video browsing of the sport activities. Game analysis of badminton as of the other racket sports requires detecting the start and end of each rally period in an automated manner. This paper describes an automatic serve scene detection method employing cubic higher-order local auto-correlation (CHLAC) and multiple regression analysis (MRA). CHLAC can extract features of postures and motions of multiple persons without segmenting and tracking each person by virtue of shift-invariance and additivity, and necessitate no prior knowledge. Then, the specific scenes, such as serve, are detected by linear regression (MRA) from the CHLAC features. To demonstrate the effectiveness of our method, the experiment was conducted on video sequences of five badminton matches captured by a single ceiling camera. The averaged precision and recall rates for the serve scene detection were 95.1% and 96.3%, respectively. Keywords—Badminton, CHLAC, MRA, Video-based motion detection

Kenji Watanabe

Papers

Molecular dynamics of STAT3 on IL-6 signaling pathway in living cells.

Logistic label propagation

Automated Service Scene Detection for Badminton Game Analysis Using CHLAC and MRA

Automated service scene detection for Badminton game analysis Using CHLAC and MRA

Robust pruning for efficient CNNs