Constellations and the unsupervised learning of graphs

In this paper, we propose a novel method for the unsupervised clustering of graphs in the context of the constellation approach to object recognition. Such method is an EM central clustering algorithm which builds prototypical graphs on the basis of fast matching with graph transformations. Our experiments, both with random graphs and in realistic situations (visual localization), show that our prototypes improve the set median graphs and also the prototypes derived from our previous incremental method. We also discuss how the method scales with a growing number of images.

Development of multidisciplinary workshops between different subjects as an innovative teaching methodology for the acquisition of professional skills of the future architect

The implantation of the new Architecture Degree and the important normative changes in the building sector imply the need to use new teaching methodologies that enhance skills and competences in order to response to the increasing requirements demanded by society to the future architect. The aim of this paper is to present, analyze and discuss the development of multidisciplinary workshops as a new teaching methodology used in several Construction subjects of the Architecture Degree in the University of Alicante. Workshops conceived with the aim to synthesize and complement the technical knowledge acquired by the students during the Degree and to enhance the skills and competencies necessary for the professional practice. With that purpose, we decided to experiment on current subjects of the degree during this academic year, by applying the requirements defined in the future Architecture Degree in a practical way, through workshops between different subjects, superposing the technical knowledge with the resolution of constructive problems in the development of an architectural project. Developing these workshops between subjects we can dissolve the traditional boundaries between different areas of the Degree. This multidisciplinary workshop methodology allows the use of all the global knowledge acquired by students during their studies and at the same time, it enhances students’ ability to communicate and discuss their ideas and solutions in public. It also increases their capacity of self-criticism, and it foments their ability to undertake learning strategies and research in an autonomous way. The used methodology is based on the development of a practical work common to several subjects of different knowledge areas within the "Technology Block" of the future Architecture Degree. Thus, students work approaching the problem in a global way discussing simultaneously with teachers from different areas. By using these new workshops we stimulate an interactive class versus a traditional lecture. Work is evaluated continuously, valuing the participative pupil´s attitude, working in groups in class time, reaching weekly objectives and stimulating the individual responsibility and positive interdependence of the pupil inside the working group. The exercises are designed to improve students’ ability to transmit their ideas and solutions in public, knowing how to discuss and defend their technical resolutions to peers and teachers (Peer Reviewing), their capacity for self-criticism and their capacity to undertake strategies and autonomous learning processes at the same time they develop a personal research into new technologies, systems and materials. Students have shown their majority preference for this teaching methodology by the multidisciplinary workshops offered in the last years, with very satisfactory academic results. In conclusion, it can be verified nowadays the viability of the introduction of new contents and new teaching methodologies necessary for the acquisition of the skills in the future Architecture Degree, through workshops between several subjects that have had a great acceptance in students and positive contrasted academic results.

A Novel Prediction Method for Early Recognition of Global Human Behaviour in Image Sequences

Human behaviour recognition has been, and still remains, a challenging problem that involves different areas of computational intelligence. The automated understanding of people activities from video sequences is an open research topic in which the computer vision and pattern recognition areas have made big efforts. In this paper, the problem is studied from a prediction point of view. We propose a novel method able to early detect behaviour using a small portion of the input, in addition to the capabilities of it to predict behaviour from new inputs. Specifically, we propose a predictive method based on a simple representation of trajectories of a person in the scene which allows a high level understanding of the global human behaviour. The representation of the trajectory is used as a descriptor of the activity of the individual. The descriptors are used as a cue of a classification stage for pattern recognition purposes. Classifiers are trained using the trajectory representation of the complete sequence. However, partial sequences are processed to evaluate the early prediction capabilities having a specific observation time of the scene. The experiments have been carried out using the three different dataset of the CAVIAR database taken into account the behaviour of an individual. Additionally, different classic classifiers have been used for experimentation in order to evaluate the robustness of the proposal. Results confirm the high accuracy of the proposal on the early recognition of people behaviours.