Semi-supervised video segmentation using tree structured graphical models

We present a novel, implementation friendly and occlusion aware semi-supervised video segmentation algorithm using tree structured graphical models, which delivers pixel labels alongwith their uncertainty estimates. Our motivation to employ supervision is to tackle a task-specific segmentation problem where the semantic objects are pre-defined by the user. The video model we propose for this problem is based on a tree structured approximation of a patch based undirected mixture model, which includes a novel time-series and a soft label Random Forest classifier participating in a feedback mechanism. We demonstrate the efficacy of our model in cutting out foreground objects and multi-class segmentation problems in lengthy and complex road scene sequences. Our results have wide applicability, including harvesting labelled video data for training discriminative models, shape/pose/articulation learning and large scale statistical analysis to develop priors for video segmentation. © 2011 IEEE.

Multiple grasp-specific representations of tool dynamics mediate skillful manipulation.

Skillful tool use requires knowledge of the dynamic properties of tools in order to specify the mapping between applied force and tool motion. Importantly, this mapping depends on the orientation of the tool in the hand. Here we investigate the representation of dynamics during skillful manipulation of a tool that can be grasped at different orientations. We ask whether the motor system uses a single general representation of dynamics for all grasp contexts or whether it uses multiple grasp-specific representations. Using a novel robotic interface, subjects rotated a virtual tool whose orientation relative to the hand could be varied. Subjects could immediately anticipate the force direction for each orientation of the tool based on its visual geometry, and, with experience, they learned to parameterize the force magnitude. Surprisingly, this parameterization of force magnitude showed limited generalization when the orientation of the tool changed. Had subjects parameterized a single general representation, full generalization would be expected. Thus, our results suggest that object dynamics are captured by multiple representations, each of which encodes the mapping associated with a specific grasp context. We suggest that the concept of grasp-specific representations may provide a unifying framework for interpreting previous results related to dynamics learning.

Raman spectroscopy as a versatile tool for studying the properties of graphene.

Raman spectroscopy is an integral part of graphene research. It is used to determine the number and orientation of layers, the quality and types of edge, and the effects of perturbations, such as electric and magnetic fields, strain, doping, disorder and functional groups. This, in turn, provides insight into all sp(2)-bonded carbon allotropes, because graphene is their fundamental building block. Here we review the state of the art, future directions and open questions in Raman spectroscopy of graphene. We describe essential physical processes whose importance has only recently been recognized, such as the various types of resonance at play, and the role of quantum interference. We update all basic concepts and notations, and propose a terminology that is able to describe any result in literature. We finally highlight the potential of Raman spectroscopy for layered materials other than graphene.

EXPERIMENTAL TECHNIQUES IN COMBUSTION USE OF IN-CYLINDER IONISATION FOR THE INVESTIGATION AND CONTROL OF SPARK IGNITION ENGINES.

Ions generated during combustion have been used in three ways to give qualitative combustion information. Langmuir type probes have been inserted into the combustion chamber opposite the spark plug location. The centre electrode of the sparking plug itself has been used to produce an ionisation signal from the slightly ionised gases remaining after the flame front has departed. The spark discharge at ignition time has been used as an anemometer.

Gaussian fading is the worst fading

The capacity of peak-power limited, single-antenna, noncoherent, flat-fading channels with memory is considered. The emphasis is on the capacity pre-log, i.e., on the limiting ratio of channel capacity to the logarithm of the signal-to-noise ratio (SNR), as the SNR tends to infinity. It is shown that, among all stationary and ergodic fading processes of a given spectral distribution function and whose law has no mass point at zero, the Gaussian process gives rise to the smallest pre-log. The assumption that the law of the fading process has no mass point at zero is essential in the sense that there exist stationary and ergodic fading processes whose law has a mass point at zero and that give rise to a smaller pre-log than the Gaussian process of equal spectral distribution function. An extension of these results to multiple-input single-output (MISO) fading channels with memory is also presented. © 2006 IEEE.

Gaussian fading is the worst fading

The capacity of peak-power limited, single-antenna, non-coherent, flat-fading channels with memory is considered. The emphasis is on the capacity pre-log, i.e., on the limiting ratio of channel capacity to the logarithm of the signal-to-noise ratio (SNR), as the SNR tends to infinity. It is shown that, among all stationary & ergodic fading processes of a given spectral distribution function whose law has no mass point at zero, the Gaussian process gives rise to the smallest pre-log. © 2006 IEEE.

Statistical user simulation with a hidden agenda

Recent work in the area of probabilistic user simulation for training statistical dialogue managers has investigated a new agenda-based user model and presented preliminary experiments with a handcrafted model parameter set. Training the model on dialogue data is an important next step, but non-trivial since the user agenda states are not observable in data and the space of possible states and state transitions is intractably large. This paper presents a summary-space mapping which greatly reduces the number of state transitions and introduces a tree-based method for representing the space of possible agenda state sequences. Treating the user agenda as a hidden variable, the forward/backward algorithm can then be successfully applied to iteratively estimate the model parameters on dialogue data. © 2007 Association for Computational Linguistics.

Superconducting fault current limiter design using parallel-connected YBCO thin films

Superconducting Fault Current Limiters (SFCLs) are able to reduce fault currents to an acceptable value, reducing potential mechanical and thermal damage to power system apparatus and allowing more flexibility in power system design and operation. The device can also help avoid replacing circuit breakers whose capacity has been exceeded. Due to limitations in current YBCO thin film manufacturing processes, it is not easy to obtain one large thin film that satisfies the specifications for high voltage and large current applications. The combination of standardized thin films has merit to reduce costs and maintain device quality, and it is necessary to connect these thin films in different series and parallel configurations in order to meet these specifications. In this paper, the design of a resistive type SFCL using parallel-connected YBCO thin films is discussed, including the role of a parallel resistor and the influence of individual thin film characteristics, based on both theory and experimental results. © 2009 IEEE.

Polarization dependence of double resonant Raman scattering band in bilayer graphene

The polarization dependence of the double resonant Raman scattering (2D) band in bilayer graphene (BLG) is studied as a function of the excitation laser energy. It has been known that the complex shape of the 2D band of BLG can be decomposed into four Lorentzian peaks with different Raman frequency shifts attributable to four individual scattering paths in the energy-momentum space. From our polarization dependence study, however, we reveal that each of the four different peaks is actually doubly degenerate in its scattering channels, i.e., two different scattering paths with similar Raman frequency shifts for each peak. We find theoretically that one of these two paths, ignored for a long time, has a small contribution to their scattering intensities but are critical in understanding their polarization dependences. Because of this, the maximum-to-minimum intensity ratios of the four peaks show a strong dependence on the excitation energy, unlike the case of single-layer graphene (SLG). Our findings thus reveal another interesting aspect of electron-phonon interactions in graphitic systems. © 2014 Elsevier Ltd. All rights reserved.