Low light situations pose a significant challenge to individuals working in a variety of different fields such as firefighting, rescue, maintenance and medicine. Tools like flashlights and infrared (IR) cameras have been used to augment light in the past, but they must often be operated manually, provide a field of view that is decoupled from the operator's own view, and utilize color schemes that can occlude content from the original scene. To help address these issues, we present VisMerge, a framework that combines a thermal imaging head mounted display (HMD) and algorithms that temporally and spectrally merge video streams of different light bands into the same field of view. For temporal synchronization, we first develop a variant of the time warping algorithm used in virtual reality (VR), but redesign it to merge video see-through (VST) cameras with different latencies. Next, using computer vision and image compositing we develop five new algorithms designed to merge non-uniform video streams from a standard RGB camera and small form-factor infrared (IR) camera. We then implement six other existing fusion methods, and conduct a series of comparative experiments, including a system level analysis of the augmented reality (AR) time warping algorithm, a pilot experiment to test perceptual consistency across all eleven merging algorithms, and an in-depth experiment on performance testing the top algorithms in a VR (simulated AR) search task. Results showed that we can reduce temporal registration error due to inter-camera latency by an average of 87.04%, that the wavelet and inverse stipple algorithms were perceptually rated the highest, that noise modulation performed best, and that freedom of user movement is significantly increased with visualizations engaged.

This paper reports the design of a push-to-talk service operated over a GPRS/UMTS network (push-to-talk over cellular - PoC). The new service was built based on the 3GPP IP multimedia subsystem (IMS). We discuss the key design decisions made for our solution and describe how the IMS supports the new service. The contributions of this paper are: [1] the verification of the overall PoC approach through an implementation; [2] a discussion of design details that were not defined in the original PoC specifications; and [3] the measurement results achieved in a live network based on existing GPRS access technology.

For part I see Hewlett Packard Technical Report HPL-97-147 (1997). Part I studied the problem of allocating resources in shared single hub 802.12 networks. We described the packet scheduling model and defined the admission control conditions which enable the network to provide deterministic service guarantees. In this paper we analyse cascaded (multihub) 802.12 topologies. We derive the relevant network parameters and show that the admission control conditions defined in part I also apply to cascaded topologies when used with these parameters. Experimental results were achieved with a UNIX kernel based implementation in cascaded test networks. These confirm the theoretical results received for network parameters, throughput and end-to-end delay.

We study the problem of allocating resources in single-hub and cascaded 802.12 networks. We show that the use of the 802.12 high-priority access mechanism, when combined with admission control, allows the network to provide small deterministic delay bounds in large cascaded network topologies with potentially many hundreds of hosts. The first part of this work is to analyze relevant 802.12 network performance parameters. In the second part, we describe the scheduling model and define the admission control conditions. Experimental results received with a UNIX kernel-based implementation in a standard 802.12 test network confirm our theoretical results for network parameters, throughput, and delay bounds. In this paper, the single-hub topology is analyzed.

Reports on the feasibility of applying neural networks to the problem of forecasting the Canada/US spot exchange rate. The inputs to the network consist of the short-term trend in the spot rate (Monday-Thursday) and the change in interest rate spread between the two countries (Wednesday-Thursday). The output is a prediction of the exchange rate on Friday. The model is able to explain 6.6% of the variance in the data between November 1991 and June 1993 (out of the sample), yielding a success rate of 59% on directional forecasts.