Abstract
A main requirement in clinical gait analysis is the ability to accurately identify gait events; especially, the initial contact of the heel with the floor and the toe off. The knowledge of the major events of the gait cycle is needed, for instance, in biomechanical data normalization and in the calculation of several temporal/distance parameters. The most common technologies for gait event detection are foot switches and force platforms; however, if the same performance could be achieved, it would be preferable to use the information collected by visual sensors to detect the main gait events. This paper proposes a procedure to be used in the detection of not just stance phase events (i.e. initial contact, opposite toe off, heel rise, opposite initial contact), but also of swing phase events (i.e. toe off, feet adjacent, tibia vertical) through the analysis of visual gait data acquired by image cameras. Moreover, in this paper, it is compared the performance of detecting the initial contact and toe off using our visual methodology and the information obtained from force platforms.

Abstract
Recent advances in mobile positioning systems and telecommunications are providing the technology needed for the development of location-aware tele-care applications. This paper introduces CAALYX - Complete Ambient Assisted Living Experiment, an EU-funded project that aims at increasing older people's autonomy and self-confidence by developing a wearable light device capable of measuring specific vital signs of the elderly, detecting falls and location, and communicating automatically in real-time with his/her care provider in case of an emergency, wherever the older person happens to be, at home or outside.

Abstract
A smart transducer is the integration of a sensor/actuator element, a processing unit, and a network interface. Smart sensor networks are composed of smart transducer nodes interconnected through a communication network. This paper presents an event driven smart sensor network architecture (asynchronous data) and its respective implementation based in the IEEE 802.15.4 standard. The events are derived from a data compression algorithm embedded into the smart sensor, which compresses data from the sensor. The architecture also supports configuration and monitoring activities for the over all distributed system.

Abstract
This paper describes a signal recognition system that is jointly optimized from mathematical representation, algorithm design and final implementation. The goal is to exploit signal properties to jointly optimize a computation, beginning with first principles (mathematical representation) and completed with implementation. We use a BestBasis algorithm to search a large collection of orthogonal transforms derived from the Walsh-Hadamard transform to find a series of transforms which best discriminate among signal classes. The implementation exploits the structure of these matrices to compress the matrix representation, and in the process of multiplying the signal by the transform, reuse the results of prior computation and parallelize the implementation in hardware. Through this joint optimization, this dynamic, data-driven system is able to yield much more highly optimized results than if the optimizations were performed statically and in isolation. We provide results taken from applying this system to real input signals of spoken digits, and perform the initial analyses to demonstrate the properties of the transform matrices lead to optimized solutions.

Abstract
Sequences of data-dependent tasks, each one traversing large data sets, exist in many applications (such as video, image and signal processing applications). Those tasks usually perform computations (with loop intensive behavior) and produce new data to be consumed by subsequent tasks. This paper shows a scheme to pipeline sequences of data-dependent loops, in such a way that subsequent loops can start execution before the completion of the previous ones, which achieves performance improvements. It uses a hardware scheme with decoupled and concurrent data-path and control units that start execution at the same time. The communication of array elements between two loops in sequence is performed by special buffers with a data-driven, fine-grained scheme. Buffer elements are responsible to flag the availability of each array element requested by a subsequent loop (i.e., a ready protocol is used to trigger the execution of operations in the succeeding loop). Thus, the control execution of following loops is also orchestrated by data availability (in this case at the array element grain) and out-of-order produced-consumed pairs are permitted. The concept has been applied using Nau, a compiler infrastructure to map algorithms described in Java onto FPGAs. This paper presents very encouraging results showing important performance improvements and buffer size reductions for a number of benchmarks.

Abstract
In this paper we present a management model to deal with the problem of tracking missing features during long image sequences using Computational Vision. Some usual difficulties related with missing features are that they may be temporarily occluded or might even have disappeared definitively, and the computational cost involved should always be reduced to the strictly necessary. The proposed Net Present Value (NPV) model, based on the economic Theory of Capital, considers the tracking of each missing feature as an investment. Thus, using the NPV criterion, with adequate receipt and outlay functions, each occluded feature may be kept on tracking or it may be excluded of the tracking process depending on its historical behavior. This approach may be applied to any tracking system as long as the tracking results may be evaluated in each temporal step, and it can deal with the appearance, occlusion and disappearance of features especially useful for tracking in long image sequences. Experimental results, both on synthetic and real image sequences, which validate our model, will be also presented.