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 increasing use of wireless networks and the constant miniaturization of electrical devices has empowered the development of Wireless Body Area Networks (WBANs). In these networks various sensors are attached on clothing or on the body or even implanted under the skin. The wireless nature of the network and the wide variety of sensors offer numerous new, practical and innovative applications to improve health care and the Quality of Life. The sensors of a WBAN measure for example the heartbeat, the body temperature or record a prolonged electrocardiogram. Using a WBAN, the patient experiences a greater physical mobility and is no longer compelled to stay in the hospital. This paper offers a survey of the concept of Wireless Body Area Networks. First, we focus on some applications with special interest in patient monitoring. Then the communication in a WBAN and its positioning between the different technologies is discussed. An overview of the current research on the physical layer, existing MAC and network protocols is given. Further, cross layer and quality of service is discussed. As WBANs are placed on the human body and often transport private data, security is also considered. An overview of current and past projects is given. Finally, the open research issues and challenges are pointed out.

/pdf/a-survey-on-wireless-body-area-networks-f3nx5ii4yv.pdf

A survey on wireless body area networks

Rapid advancement of technology continues to leverage change and innovation in the construction industry. Continued digitization of the industry offers the opportunity to totally reinvent contemporary construction design and delivery practice for future development. Building Information Modelling (BIM) within the context of Architecture, Engineering & Construction (AEC) has been developing since the early 2000s and is considered to be a key technology. Despite major technical advancements in BIM, it has not been fully adopted and its definitive benefits have not been fully capitalized upon by industry stakeholders. The lack of widespread uptake of BIM appears to be linked to the risks and challenges that are potentially impeding its effectiveness. This paper aims to discuss the reality of BIM, its widespread benefits and current level of uptake. The risks and challenges associated with the adoption of BIM, as well as recommendations regarding how future BIM adoption could be developed are also highlighted.

Building Information Modelling (BIM) uptake: Clear benefits, understanding its implementation, risks and challenges

Time series shapelets are a recent promising concept in time series data mining. Shapelets are time series snippets that can be used to classify unlabeled time series. Shapelets not only provide interpretable results, which are useful for domain experts and developers alike, but shapelet-based classifiers have been shown by several independent research groups to have superior accuracy on many datasets. Moreover, shapelets can be seen as generalizing the lazy nearest neighbor classifier to an eager classifier. Thus, as a deployed classification tool, shapelets can be many orders of magnitude faster than any rival with comparable accuracy. Although shapelets are a useful concept, the current literature bemoans the fact that shapelet discovery is a time-consuming task. In spite of several efforts to speed up shapelet discovery algorithms, including the use of specialist hardware, the current state-of-the-art algorithms are still intractable on large datasets. In this work, we propose a fast shapelet discovery algorithm that outperforms the current state-of-the-art by two or three orders of magnitude, while producing models with accuracy that is not perceptibly different.

/pdf/fast-shapelets-a-scalable-algorithm-for-discovering-time-1gmlsbxf5l.pdf

Fast Shapelets: A Scalable Algorithm for Discovering Time Series Shapelets.

In this paper, we examine the main advances registered in the last ten years in Pattern Recognition methodologies based on graph matching and related techniques, analyzing more than 180 papers; the...

Graph matching and learning in pattern recognition in the last 10 years

This paper proposes a novel complete system for automated floor plan analysis. Besides applying and improving state-of-the-art processing methods, we introduce novel preprocessing methods, e.g., the differentiation between thick, medium, and thin lines and the removal of components outside the convex hull of the outer walls. Especially the latter method increases the performance of the final system. In our experiments on a reference data set we compare our approach to other approaches available in the literature. We show that our system outperforms previous systems. The final room recognition accuracy is 79 % that is 10 % higher than the 69 % achieved by a state-of-the-art approach from the literature.

Improved Automatic Analysis of Architectural Floor Plans

This paper presents an automatic system for analyzing and labeling architectural floor plans. In order to detect the locations of the rooms, the proposed systems extracts both, structural and semantic information from given floor plans. Furthermore, OCR is applied on the text layer to retrieve the meaningful room labeling. Finally, a novel post-processing is proposed to split rooms into several sub-regions if several semantic rooms share the same physical room. Our fully automatic system is evaluated on a publicly available dataset of architectural floor plans. In our experiments, we could clearly outperform other state-of-the-art approaches for room detection.

/pdf/automatic-room-detection-and-room-labeling-from-40zk4lyd4g.pdf

Automatic Room Detection and Room Labeling from Architectural Floor Plans

Graph-based retrieval of building information models for supporting the early design stages

Automatic analysis and sketch-based retrieval of architectural floor plans

Regular and moderate physical activity practice provides many physiological benefits. It reduces the risk of disease outcomes and is the basis for proper rehabilitation after a severe disease. Aerobic activity and strength exercises are strongly recommended in order to maintain autonomy with ageing. Balanced activity of both types is important, especially to the elderly population. Several methods have been proposed to monitor aerobic activities. However, no appropriate method is available for controlling more complex parameters of strength exercises. Within this context, the present article introduces a personalized, home-based strength exercise trainer designed for the elderly. The system guides a user at home through a personalized exercise program. Using a network of wearable sensors the user's motions are captured. These are evaluated by comparing them to prescribed exercises, taking both exercise load and technique into account. Moreover, the evaluation results are immediately translated into appropriate feedback to the user in order to assist the correct exercise execution. Besides the direct feedback, a major novelty of the system is its generic personalization by means of a supervised teach-in phase, where the program is performed once under supervision of a physical activity specialist. This teach-in phase allows the system to record and learn the correct execution of exercises for the individual user and to provide personalized monitoring. The user-driven design process, the system development and its underlying activity monitoring methodology are described. Moreover, technical evaluation results as well as results concerning the usability of the system for ageing people are presented. The latter has been assessed in a clinical study with thirty participants of 60 years or older, some of them showing usual diseases or functional limitations observed in elderly population.

https://hal-univ-rennes1.archives-ouvertes.fr/hal-00934916/document

Markus Weber

Papers

Improved Automatic Analysis of Architectural Floor Plans

Automatic Room Detection and Room Labeling from Architectural Floor Plans

Graph-based retrieval of building information models for supporting the early design stages

Automatic analysis and sketch-based retrieval of architectural floor plans

A personalized exercise trainer for the elderly