Search

This study investigates the accuracy of the tracking system LPM (local position measurement). The goal was to determine detailed error values of the system in the context of sports performance analyses. Six moderately trained male soccer players (amateur level) performed 276 runs on three different courses at six different speeds. Additionally, ten small-sided game plays were carried out. All runs and game plays were recorded with the LPM tracking system and the motion capture system VICON simultaneously. VICON served as the reference system. The absolute error of all LPM position estimations was on average 23.4±20.7 cm. The estimation for average velocities varied between 0.01 km h(-1) and 0.23 km h(-1), the maximum speed estimations differed by up to 2.71 km h(-1). In addition, the results showed that the accuracy of the LPM system is highly dependent on the instantaneous dynamics of the player and decreases in the margins of the observation field. These dependencies were quantified. Considering commonly used applications of position tracking systems in sports (Leser, Ogris,Baca, 2011), the accuracy of LPM is acceptable for position and velocity estimations. The system provides valuable results for average velocities but seems to be far less reliable when dealing with high dynamic movements and measuring instantaneous velocities.

We describe the design and evaluation of pattern analysis methods for the recognition of maintenance-related activities. The presented work focuses on the spotting of subtle hand actions in a continuous stream of data based on a combination of body-mounted motion sensors and ultrasonic positioning. The spotting and recognition approach is based on three core ideas: (1) the use of location information to compensate for the ambiguity of hand motions, (2) the use of motion data to compensate for the slow sampling rate and unreliable signal of the low cost ultrasonic positioning system, and (3) an incremental, multistage spotting methodology. The proposed methods are evaluated in an elaborate bicycle repair experiment containing nearly 10 h of data from six subjects. The evaluation compares different strategies and system variants and shows that precision and recall rates around 90% can be achieved.

Position data of players and athletes are widely used in sports performance analysis for measuring the amounts of physical activities as well as for tactical assessments in game sports. However, positioning sensing systems are applied in sports as tools to gain objective information of sports behavior rather than as components of intelligent spaces (IS). The paper outlines the idea of IS for the sports context with special focus to game sports and how intelligent sports feedback systems can benefit from IS. Henceforth, the most common location sensing techniques used in sports and their practical application are reviewed, as location is among the most important enabling techniques for IS. Furthermore, the article exemplifies the idea of IS in sports on two applications.

We present a novel method for continuous activity recognition based on ultrasonic hand tracking and motion sensors attached to the user's arms. It builds on previous work in which we have shown such a sensor combination to be effective for isolated recognition in manually segmented data. We describe the hand tracking based segmentation, show how classification is done on both the ultrasonic and the motion data and discuss different classifier fusion methods. The performance of our method is investigated in a large scale experiment in which typical bicycle repair actions are performed by 6 different subjects. The experiment contains a test set with 1008 activities from 21 classes encompassing 115 minutes randomly mixed with 252 minutes of 'NULL' class. To come as close as possible to a real life continuous scenario we have ensured a diverse and complex 'NULL' class, diverse and often similar activities, inter person training/testing and an additional data set only for training (299 extra minutes of data). A key result of the paper is that our method can handle the user independent testing (testing on users that were not seen in training) nearly as well as the user dependent case.

The paper demonstrates how ultrasonic hand tracking can be used to improve the performance of a wearable, accelerometer and gyroscope based activity recognition system. Specifically we target the recognition of manipulative gestures of the type found in assembly and maintenance tasks. We discuss how relevant information can be extracted from the ultrasonic signal despite problems with low sampling rate, occlusions and reflections that occur in this type of application. We then introduce several methods of fusing the ultrasound and motion sensor information. We evaluate our methods on an experimental data set that contains 21 different actions performed repeatedly by three different subjects during simulated bike repair. Due to the complexity of the recognition tasks with many similar and vaguely defined actions and person independent training both the ultrasound and motion sensors perform poorly on their own. However with our fusion methods recognition rates well over 90% can be achieved for most activities. In extreme case recognition rates go up from just over 50% for separate classifications to nearly 89% with our fusion methods.

This work is conceptually based on psychological findings in human perception that highlight the utility of scene interpretation in object detection processes. Objects of interest are embedded in their visual context, i.e., in visual events within their spatial neighborhood. The implication for a detection system is that early recognition of this environment might provide information to directly map to an object event. The original contribution of this work is to outline a detection system that gains prospective information out of rapid scene analysis in order to focus attention on estimated object locations. Scene recognition is outlined on the basis of rapid detection of triplet configurations of landmarks which determine the discriminability of a particular location within the scene. Formulating scene recognition in terms of posterior landmark interpretation enables a recursive integration of target predictions and hence a probabilistic representation for attention based object detection.

Movement variability has received great attention due to evidence of its functional meaning, and thus offers potential for training optimization. The study on hand presents an extended pedestrian-dead-reckoning approach to stride parameter estimation in running on an athlete track. In addition, the supposed measurement system is evaluated by means of a video-based reference system. This pilot study revealed that the proposed approach promises to be successful for high accuracy stride parameter estimation.