Learning object dynamics for smooth tracking of moving lip contours

A new technique is proposed for learning the dynamic characteristics of a deformable object, applied in particular to the problem of lip-tracking. Experimental results are given which demonstrate that the use of dynamic models allows the system to track more robustly under adverse conditions and to correct spurious, poorly tracked frames

Empirical comparison of machine learning algorithms for image texture classification with application to vegetation management in power line corridors

This paper reports on the empirical comparison of seven machine learning algorithms in texture classification with application to vegetation management in power line corridors. Aiming at classifying tree species in power line corridors, object-based method is employed. Individual tree crowns are segmented as the basic classification units and three classic texture features are extracted as the input to the classification algorithms. Several widely used performance metrics are used to evaluate the classification algorithms. The experimental results demonstrate that the classification performance depends on the performance matrix, the characteristics of datasets and the feature used.

Path information effects in visual and proprioceptive spatial learning

Objects in an environment are often encountered sequentially during spatial learning, forming a path along which object locations are experienced. The present study investigated the effect of spatial information conveyed through the path in visual and proprioceptive learning of a room-sized spatial layout, exploring whether different modalities differentially depend on the integrity of the path. Learning object locations along a coherent path was compared with learning them in a spatially random manner. Path integrity had little effect on visual learning, whereas learning with the coherent path produced better memory performance than random order learning for proprioceptive learning. These results suggest that path information has differential effects in visual and proprioceptive spatial learning, perhaps due to a difference in the way one establishes a reference frame for representing relative locations of objects.

Sequential versus simultaneous viewing of an environment : effects of focal attention to individual object locations on visual spatial learning

We investigated memories of room-sized spatial layouts learned by sequentially or simultaneously viewing objects from a stationary position. In three experiments, sequential viewing (one or two objects at a time) yielded subsequent memory performance that was equivalent or superior to simultaneous viewing of all objects, even though sequential viewing lacked direct access to the entire layout. This finding was replicated by replacing sequential viewing with directed viewing in which all objects were presented simultaneously and participants’ attention was externally focused on each object sequentially, indicating that the advantage of sequential viewing over simultaneous viewing may have originated from focal attention to individual object locations. These results suggest that memory representation of object-to-object relations can be constructed efficiently by encoding each object location separately, when those locations are defined within a single spatial reference system. These findings highlight the importance of considering object presentation procedures when studying spatial learning mechanisms.

Mid-level concept learning with visual contextual ontologies and probabilistic inference for image annotation

To date, automatic recognition of semantic information such as salient objects and mid-level concepts from images is a challenging task. Since real-world objects tend to exist in a context within their environment, the computer vision researchers have increasingly incorporated contextual information for improving object recognition. In this paper, we present a method to build a visual contextual ontology from salient objects descriptions for image annotation. The ontologies include not only partOf/kindOf relations, but also spatial and co-occurrence relations. A two-step image annotation algorithm is also proposed based on ontology relations and probabilistic inference. Different from most of the existing work, we specially exploit how to combine representation of ontology, contextual knowledge and probabilistic inference. The experiments show that image annotation results are improved in the LabelMe dataset.

Transfer of learning across the globe : pre-service teachers taking the initiative

A continuing challenge for pre-service teacher education is the learning transfer between the university based components and the practical school based components of their training. It is not clear how easily pre-service teachers can transfer university learnings into ‘in school’ practice. Similarly, it is not clear how easily knowledge learned in the school context can be disembedded from this particular context and understood more generally by the pre-service teacher. This paper examines the effect of a community of practice formed specifically to explore learning transfer via collaboration and professional enquiry, in ‘real time’, across the globe. “Activity Theory” (Engestrom, 1999) provided the theoretical framework through which the cognitive, physical and social processes involved could be understood. For the study, three activity systems formed community of practice network. The first activity system involved pre-service teachers at a large university in Queensland, Australia. The second activity system was introduced by the pre-service teachers and involved Year 12 students and teachers at a private secondary school also in Queensland, Australia. The third activity system involved university staff engineers at a large university in Pennsylvania, USA. The common object among the three activity systems was to explore the principles and applications of nanotechnology. The participants in the two Queensland activity systems, controlled laboratory equipment (a high powered Atomic Force Microscope – CPII) in Pennsylvania, USA, with the aim of investigating surface topography and the properties of nano particles. The pre-service teachers were to develop their remote ‘real time’ experience into school classroom tasks, implement these tasks, and later report their findings to other pre-service teachers in the university activity system. As an extension to the project, the pre-service teachers were invited to co-author papers relating to the project. Data were collected from (a) reflective journals; (b) participant field notes – a pre-service teacher initiative; (c) surveys – a pre-service teacher initiative; (d) lesson reflections and digital recordings – a pre-service teacher initiative; and (e) interviews with participants. The findings are reported in terms of the major themes: boundary crossing, the philosophy of teaching, and professional relationships The findings have implications for teacher education. The researchers feel that deliberate planning for networking between activity systems may well be a solution to the apparent theory/practice gap. Proximity of activity systems need not be a hindering issue.

Learning articulated motion structures with Bayesian Networks

This paper presents a general methodology for learning articulated motions that, despite having non-linear correlations, are cyclical and have a defined pattern of behavior Using conventional algorithms to extract features from images, a Bayesian classifier is applied to cluster and classify features of the moving object. Clusters are then associated in different frames and structure learning algorithms for Bayesian networks are used to recover the structure of the motion. This framework is applied to the human gait analysis and tracking but applications include any coordinated movement such as multi-robots behavior analysis.

Applying incremental EM to Bayesian classifiers in the learning of hyperspectral remote sensing data

In this paper, we apply the incremental EM method to Bayesian Network Classifiers to learn and interpret hyperspectral sensor data in robotic planetary missions. Hyperspectral image spectroscopy is an emerging technique for geological investigations from airborne or orbital sensors. Many spacecraft carry spectroscopic equipment as wavelengths outside the visible light in the electromagnetic spectrum give much greater information about an object. The algorithm used is an extension to the standard Expectation Maximisation (EM). The incremental method allows us to learn and interpret the data as they become available. Two Bayesian network classifiers were tested: the Naive Bayes, and the Tree-Augmented-Naive Bayes structures. Our preliminary experiments show that incremental learning with unlabelled data can improve the accuracy of the classifier.

Emergent task-specific object semantics through distributed experience networks

Autonomous development of sensorimotor coordination enables a robot to adapt and change its action choices to interact with the world throughout its lifetime. The Experience Network is a structure that rapidly learns coordination between visual and haptic inputs and motor action. This paper presents methods which handle the high dimensionality of the network state-space which occurs due to the simultaneous detection of multiple sensory features. The methods provide no significant increase in the complexity of the underlying representations and also allow emergent, task-specific, semantic information to inform action selection. Experimental results show rapid learning in a real robot, beginning with no sensorimotor mappings, to a mobile robot capable of wall avoidance and target acquisition.

The activity of learning assistance

This paper uses cultural-historical activity to examine the nature of learning assistance, provided in university settings, to assist learners in coping with academic study. Alternative rival constructions of learning assistance, either as part of an overall university activity system, or as an activity system overlapping with other separate activity systems (e.g. library and faculty) within the university, are outlined. Data from a research study at one university are used to describe the object, tools and tensions in learning assistance; and, together with the history of the service, these are used to suggest that the current situation is one where the objects of the Learning Assistance Unit, the faculty and the library have separated from the object of a university.

Learning landscapes 2.0 : warning - assembly required

Twenty first century learners operate in organic, immersive environments. A pedagogy of student-centred learning is not a recipe for rooms. A contemporary learning environment is like a landscape that grows, morphs, and responds to the pressures of the context and micro-culture. There is no single adaptable solution, nor a suite of off-the-shelf answers; propositions must be customisable and infinitely variable. They must be indeterminate and changeable; based on the creation of learning places, not restrictive or constraining spaces. A sustainable solution will be un-fixed, responsive to the life cycle of the components and materials, able to be manipulated by the users; it will create and construct its own history. Learning occurs as formal education with situational knowledge structures, but also as informal learning, active learning, blended learning social learning, incidental learning, and unintended learning. These are not spatial concepts but socio-cultural patterns of discovery. Individual learning requirements must run free and need to be accommodated as the learner sees fit. The spatial solution must accommodate and enable a full array of learning situations. It is a system not an object. Three major components: 1. The determinate landscape: in-situ concrete 'plate' that is permanent. It predates the other components of the system and remains as a remnant/imprint/fossil after the other components of the system have been relocated. It is a functional learning landscape in its own right; enabling a variety of experiences and activities. 2. The indeterminate landscape: a kit of pre-fabricated 2-D panels assembled in a unique manner at each site to suit the client and context. Manufactured to the principles of design-for-disassembly. A symbiotic barnacle like system that attaches itself to the existing infrastructure through the determinate landscape which acts as a fast growth rhizome. A carapace of protective panels, infinitely variable to create enclosed, semi-enclosed, and open learning places. 3. The stations: pre-fabricated packages of highly-serviced space connected through the determinate landscape. Four main types of stations; wet-room learning centres, dry-room learning centres, ablutions, and low-impact building services. Entirely customised at the factory and delivered to site. The stations can be retro-fitted to suit a new context during relocation. Principles of design for disassembly: material principles • use recycled and recyclable materials • minimise the number of types of materials • no toxic materials • use lightweight materials • avoid secondary finishes • provide identification of material types component principles • minimise/standardise the number of types of components • use mechanical not chemical connections • design for use of common tools and equipment • provide easy access to all components • make component size to suite means of handling • provide built in means of handling • design to realistic tolerances • use a minimum number of connectors and a minimum number of types system principles • design for durability and repeated use • use prefabrication and mass production • provide spare components on site • sustain all assembly and material information

Robots move : bootstrapping the development of object representations using sensorimotor coordination

This paper is concerned with the unsupervised learning of object representations by fusing visual and motor information. The problem is posed for a mobile robot that develops its representations as it incrementally gathers data. The scenario is problematic as the robot only has limited information at each time step with which it must generate and update its representations. Object representations are refined as multiple instances of sensory data are presented; however, it is uncertain whether two data instances are synonymous with the same object. This process can easily diverge from stability. The premise of the presented work is that a robot's motor information instigates successful generation of visual representations. An understanding of self-motion enables a prediction to be made before performing an action, resulting in a stronger belief of data association. The system is implemented as a data-driven partially observable semi-Markov decision process. Object representations are formed as the process's hidden states and are coordinated with motor commands through state transitions. Experiments show the prediction process is essential in enabling the unsupervised learning method to converge to a solution - improving precision and recall over using sensory data alone.

Locally weighted learning model predictive control for nonlinear and time varying dynamics

This paper proposes an online learning control system that uses the strategy of Model Predictive Control (MPC) in a model based locally weighted learning framework. The new approach, named Locally Weighted Learning Model Predictive Control (LWL-MPC), is proposed as a solution to learn to control robotic systems with nonlinear and time varying dynamics. This paper demonstrates the capability of LWL-MPC to perform online learning while controlling the joint trajectories of a low cost, three degree of freedom elastic joint robot. The learning performance is investigated in both an initial learning phase, and when the system dynamics change due to a heavy object added to the tool point. The experiment on the real elastic joint robot is presented and LWL-MPC is shown to successfully learn to control the system with and without the object. The results highlight the capability of the learning control system to accommodate the lack of mechanical consistency and linearity in a low cost robot arm.