A Multi-View Learning Algorithm

Report published in the Proceedings of the National Conference on "Education and Research in the Information Society", Plovdiv, May, 2014

Learning and Exploiting Camera Geometry for Computer Vision

This work explores the use of statistical methods in describing and estimating camera poses, as well as the information feedback loop between camera pose and object detection. Surging development in robotics and computer vision has pushed the need for algorithms that infer, understand, and utilize information about the position and orientation of the sensor platforms when observing and/or interacting with their environment.

The first contribution of this thesis is the development of a set of statistical tools for representing and estimating the uncertainty in object poses. A distribution for representing the joint uncertainty over multiple object positions and orientations is described, called the mirrored normal-Bingham distribution. This distribution generalizes both the normal distribution in Euclidean space, and the Bingham distribution on the unit hypersphere. It is shown to inherit many of the convenient properties of these special cases: it is the maximum-entropy distribution with fixed second moment, and there is a generalized Laplace approximation whose result is the mirrored normal-Bingham distribution. This distribution and approximation method are demonstrated by deriving the analytical approximation to the wrapped-normal distribution. Further, it is shown how these tools can be used to represent the uncertainty in the result of a bundle adjustment problem.

Another application of these methods is illustrated as part of a novel camera pose estimation algorithm based on object detections. The autocalibration task is formulated as a bundle adjustment problem using prior distributions over the 3D points to enforce the objects' structure and their relationship with the scene geometry. This framework is very flexible and enables the use of off-the-shelf computational tools to solve specialized autocalibration problems. Its performance is evaluated using a pedestrian detector to provide head and foot location observations, and it proves much faster and potentially more accurate than existing methods.

Finally, the information feedback loop between object detection and camera pose estimation is closed by utilizing camera pose information to improve object detection in scenarios with significant perspective warping. Methods are presented that allow the inverse perspective mapping traditionally applied to images to be applied instead to features computed from those images. For the special case of HOG-like features, which are used by many modern object detection systems, these methods are shown to provide substantial performance benefits over unadapted detectors while achieving real-time frame rates, orders of magnitude faster than comparable image warping methods.

The statistical tools and algorithms presented here are especially promising for mobile cameras, providing the ability to autocalibrate and adapt to the camera pose in real time. In addition, these methods have wide-ranging potential applications in diverse areas of computer vision, robotics, and imaging.

AR-based Technoself Enhanced Learning Approach to Improving Student Engagement

The emerging technologies have expanded a new dimension of self – ‘technoself’ driven by socio-technical innovations and taken an important step forward in pervasive learning. Technology Enhanced Learning (TEL) research has increasingly focused on emergent technologies such as Augmented Reality (AR) for augmented learning, mobile learning, and game-based learning in order to improve self-motivation and self-engagement of the learners in enriched multimodal learning environments. These researches take advantage of technological innovations in hardware and software across different platforms and devices including tablets, phoneblets and even game consoles and their increasing popularity for pervasive learning with the significant development of personalization processes which place the student at the center of the learning process. In particular, augmented reality (AR) research has matured to a level to facilitate augmented learning, which is defined as an on-demand learning technique where the learning environment adapts to the needs and inputs from learners. In this paper we firstly study the role of Technology Acceptance Model (TAM) which is one of the most influential theories applied in TEL on how learners come to accept and use a new technology. Then we present the design methodology of the technoself approach for pervasive learning and introduce technoself enhanced learning as a novel pedagogical model to improve student engagement by shaping personal learning focus and setting. Furthermore we describe the design and development of an AR-based interactive digital interpretation system for augmented learning and discuss key features. By incorporating mobiles, game simulation, voice recognition, and multimodal interaction through Augmented Reality, the learning contents can be geared toward learner's needs and learners can stimulate discovery and gain greater understanding. The system demonstrates that Augmented Reality can provide rich contextual learning environment and contents tailored for individuals. Augment learning via AR can bridge this gap between the theoretical learning and practical learning, and focus on how the real and virtual can be combined together to fulfill different learning objectives, requirements, and even environments. Finally, we validate and evaluate the AR-based technoself enhanced learning approach to enhancing the student motivation and engagement in the learning process through experimental learning practices. It shows that Augmented Reality is well aligned with constructive learning strategies, as learners can control their own learning and manipulate objects that are not real in augmented environment to derive and acquire understanding and knowledge in a broad diversity of learning practices including constructive activities and analytical activities.

Children with Learning Disabilities and Challenging Behaviour: Relationship between Parents and Professionals

This multi-perspectival Interpretive Phenomenological Analysis (IPA) study explored how people in the ‘networks of concern’ talked about how they tried to make sense of the challenging behaviours of four children with severe learning disabilities. The study also aimed to explore what affected relationships between people. The study focussed on 4 children through interviewing their mothers, their teachers and the Camhs Learning Disability team members who were working with them. Two fathers also joined part of the interviews. All interviews were conducted separately using a semi-structured approach. IPA allowed both a consideration of the participant’s lived experiences and ‘objects of concern’ and a deconstruction of the multiple contexts of people’s lives, with a particular focus on disability. The analysis rendered five themes: the importance of love and affection, the difficulties, and the differences of living with a challenging child, the importance of being able to make sense of the challenges and the value of good relationships between people. Findings were interpreted through the lens of CMM (Coordinated Management of Meaning), which facilitated a systemic deconstruction and reconstruction of the findings. The research found that making sense of the challenges was a key concern for parents. Sharing meanings were important for people’s relationships with each other, including employing diagnostic and behavioural narratives. The importance of context is also highlighted including a consideration of how societal views of disability have an influence on people in the ‘network of concern’ around the child. A range of systemic approaches, methods and techniques are suggested as one way of improving services to these children and their families. It is suggested that adopting a ‘both/and’ position is important in such work - both applying evidence based approaches and being alert to and exploring the different ways people try and make sense of the children’s challenges. Implications for practice included helping professionals be alert to their constructions and professional narratives, slowing the pace with families, staying close to the concerns of families and addressing network issues.

Learning Physics with the Unreal Tournament Engine

Computer games such as Unreal Tournament (UT2004 and UT3) contain a 'physics engine' responsible for producing believable dynamic interactions between players and objects in the three-dimensional (3D) virtual world of a game. Through a series of probing experiments we have evaluated the fidelity and internal consistency of the UT2004 physics engine. These experiments have then led to the production of resources which may be used by learners and teachers of secondary-school physics. We also suggest an approach to learning, where both teachers and pupils may produce learning materials using the Unreal Tournament editor 'UnrealEd'.

Applications of a Collaborative Open Object Oriented Learning Environment to Electronics, Computing and Mathematics Education

A lightweight Java application suite has been developed and deployed allowing collaborative learning between students and tutors at remote locations. Students can engage in group activities online and also collaborate with tutors. A generic Java framework has been developed and applied to electronics, computing and mathematics education. The applications are respectively: (a) a digital circuit simulator, which allows students to collaborate in building simple or complex electronic circuits; (b) a Java programming environment where the paradigm is behavioural-based robotics, and (c) a differential equation solver useful in modelling of any complex and nonlinear dynamic system. Each student sees a common shared window on which may be added text or graphical objects and which can then be shared online. A built-in chat room supports collaborative dialogue. Students can work either in collaborative groups or else in teams as directed by the tutor. This paper summarises the technical architecture of the system as well as the pedagogical implications of the suite. A report of student evaluation is also presented distilled from use over a period of twelve months. We intend this suite to facilitate learning between groups at one or many institutions and to facilitate international collaboration. We also intend to use the suite as a tool to research the establishment and behaviour of collaborative learning groups. We shall make our software freely available to interested researchers.

SMILE: Simulator for Maryland Imitation Learning Environment

As robot imitation learning is beginning to replace conventional hand-coded approaches in programming robot behaviors, much work is focusing on learning from the actions of demonstrators. We hypothesize that in many situations, procedural tasks can be learned more effectively by observing object behaviors while completely ignoring the demonstrator's motions. To support studying this hypothesis and robot imitation learning in general, we built a software system named SMILE that is a simulated 3D environment. In this virtual environment, both a simulated robot and a user-controlled demonstrator can manipulate various objects on a tabletop. The demonstrator is not embodied in SMILE, and therefore a recorded demonstration appears as if the objects move on their own. In addition to recording demonstrations, SMILE also allows programing the simulated robot via Matlab scripts, as well as creating highly customizable objects for task scenarios via XML. This report describes the features and usages of SMILE.

Datasets, features, learning, and models in visual recognition

Visual recognition is a fundamental research topic in computer vision. This dissertation explores datasets, features, learning, and models used for visual recognition. In order to train visual models and evaluate different recognition algorithms, this dissertation develops an approach to collect object image datasets on web pages using an analysis of text around the image and of image appearance. This method exploits established online knowledge resources (Wikipedia pages for text; Flickr and Caltech data sets for images). The resources provide rich text and object appearance information. This dissertation describes results on two datasets. The first is Berg’s collection of 10 animal categories; on this dataset, we significantly outperform previous approaches. On an additional set of 5 categories, experimental results show the effectiveness of the method. Images are represented as features for visual recognition. This dissertation introduces a text-based image feature and demonstrates that it consistently improves performance on hard object classification problems. The feature is built using an auxiliary dataset of images annotated with tags, downloaded from the Internet. Image tags are noisy. The method obtains the text features of an unannotated image from the tags of its k-nearest neighbors in this auxiliary collection. A visual classifier presented with an object viewed under novel circumstances (say, a new viewing direction) must rely on its visual examples. This text feature may not change, because the auxiliary dataset likely contains a similar picture. While the tags associated with images are noisy, they are more stable when appearance changes. The performance of this feature is tested using PASCAL VOC 2006 and 2007 datasets. This feature performs well; it consistently improves the performance of visual object classifiers, and is particularly effective when the training dataset is small. With more and more collected training data, computational cost becomes a bottleneck, especially when training sophisticated classifiers such as kernelized SVM. This dissertation proposes a fast training algorithm called Stochastic Intersection Kernel Machine (SIKMA). This proposed training method will be useful for many vision problems, as it can produce a kernel classifier that is more accurate than a linear classifier, and can be trained on tens of thousands of examples in two minutes. It processes training examples one by one in a sequence, so memory cost is no longer the bottleneck to process large scale datasets. This dissertation applies this approach to train classifiers of Flickr groups with many group training examples. The resulting Flickr group prediction scores can be used to measure image similarity between two images. Experimental results on the Corel dataset and a PASCAL VOC dataset show the learned Flickr features perform better on image matching, retrieval, and classification than conventional visual features. Visual models are usually trained to best separate positive and negative training examples. However, when recognizing a large number of object categories, there may not be enough training examples for most objects, due to the intrinsic long-tailed distribution of objects in the real world. This dissertation proposes an approach to use comparative object similarity. The key insight is that, given a set of object categories which are similar and a set of categories which are dissimilar, a good object model should respond more strongly to examples from similar categories than to examples from dissimilar categories. This dissertation develops a regularized kernel machine algorithm to use this category dependent similarity regularization. Experiments on hundreds of categories show that our method can make significant improvement for categories with few or even no positive examples.

A machine learning approach for recognising woody plants on railway trackbeds

The purpose of this work in progress study was to test the concept of recognising plants using images acquired by image sensors in a controlled noise-free environment. The presence of vegetation on railway trackbeds and embankments presents potential problems. Woody plants (e.g. Scots pine, Norway spruce and birch) often establish themselves on railway trackbeds. This may cause problems because legal herbicides are not effective in controlling them; this is particularly the case for conifers. Thus, if maintenance administrators knew the spatial position of plants along the railway system, it may be feasible to mechanically harvest them. Primary data were collected outdoors comprising around 700 leaves and conifer seedlings from 11 species. These were then photographed in a laboratory environment. In order to classify the species in the acquired image set, a machine learning approach known as Bag-of-Features (BoF) was chosen. Irrespective of the chosen type of feature extraction and classifier, the ability to classify a previously unseen plant correctly was greater than 85%. The maintenance planning of vegetation control could be improved if plants were recognised and localised. It may be feasible to mechanically harvest them (in particular, woody plants). In addition, listed endangered species growing on the trackbeds can be avoided. Both cases are likely to reduce the amount of herbicides, which often is in the interest of public opinion. Bearing in mind that natural objects like plants are often more heterogeneous within their own class rather than outside it, the results do indeed present a stable classification performance, which is a sound prerequisite in order to later take the next step to include a natural background. Where relevant, species can also be listed under the Endangered Species Act.

The three-dimensional single-bin-size bin packing problem: combining metaheuristic and machine learning approaches

The Three-Dimensional Single-Bin-Size Bin Packing Problem is one of the most studied problem in the Cutting & Packing category. From a strictly mathematical point of view, it consists of packing a finite set of strongly heterogeneous “small” boxes, called items, into a finite set of identical “large” rectangles, called bins, minimizing the unused volume and requiring that the items are packed without overlapping. The great interest is mainly due to the number of real-world applications in which it arises, such as pallet and container loading, cutting objects out of a piece of material and packaging design. Depending on these real-world applications, more objective functions and more practical constraints could be needed. After a brief discussion about the real-world applications of the problem and a exhaustive literature review, the design of a two-stage algorithm to solve the aforementioned problem is presented. The algorithm must be able to provide the spatial coordinates of the placed boxes vertices and also the optimal boxes input sequence, while guaranteeing geometric, stability, fragility constraints and a reduced computational time. Due to NP-hard complexity of this type of combinatorial problems, a fusion of metaheuristic and machine learning techniques is adopted. In particular, a hybrid genetic algorithm coupled with a feedforward neural network is used. In the first stage, a rich dataset is created starting from a set of real input instances provided by an industrial company and the feedforward neural network is trained on it. After its training, given a new input instance, the hybrid genetic algorithm is able to run using the neural network output as input parameter vector, providing as output the optimal solution. The effectiveness of the proposed works is confirmed via several experimental tests.

Robotic Manipulation of DLOs for Wiring Harness Assembly: a Machine Learning Approach

In recent times, a significant research effort has been focused on how deformable linear objects (DLOs) can be manipulated for real world applications such as assembly of wiring harnesses for the automotive and aerospace sector. This represents an open topic because of the difficulties in modelling accurately the behaviour of these objects and simulate a task involving their manipulation, considering a variety of different scenarios. These problems have led to the development of data-driven techniques in which machine learning techniques are exploited to obtain reliable solutions. However, this approach makes the solution difficult to be extended, since the learning must be replicated almost from scratch as the scenario changes. It follows that some model-based methodology must be introduced to generalize the results and reduce the training effort accordingly. The objective of this thesis is to develop a solution for the DLOs manipulation to assemble a wiring harness for the automotive sector based on adaptation of a base trajectory set by means of reinforcement learning methods. The idea is to create a trajectory planning software capable of solving the proposed task, reducing where possible the learning time, which is done in real time, but at the same time presenting suitable performance and reliability. The solution has been implemented on a collaborative 7-DOFs Panda robot at the Laboratory of Automation and Robotics of the University of Bologna. Experimental results are reported showing how the robot is capable of optimizing the manipulation of the DLOs gaining experience along the task repetition, but showing at the same time a high success rate from the very beginning of the learning phase.

Deep learning based style transfer for low altitude aerial imagery

Unmanned Aerial Vehicle (UAVs) equipped with cameras have been fast deployed to a wide range of applications, such as smart cities, agriculture or search and rescue applications. Even though UAV datasets exist, the amount of open and quality UAV datasets is limited. So far, we want to overcome this lack of high quality annotation data by developing a simulation framework for a parametric generation of synthetic data. The framework accepts input via a serializable format. The input specifies which environment preset is used, the objects to be placed in the environment along with their position and orientation as well as additional information such as object color and size. The result is an environment that is able to produce UAV typical data: RGB image from the UAVs camera, altitude, roll, pitch and yawn of the UAV. Beyond the image generation process, we improve the resulting image data photorealism by using Synthetic-To-Real transfer learning methods. Transfer learning focuses on storing knowledge gained while solving one problem and applying it to a different - although related - problem. This approach has been widely researched in other affine fields and results demonstrate it to be an interesing area to investigate. Since simulated images are easy to create and synthetic-to-real translation has shown good quality results, we are able to generate pseudo-realistic images. Furthermore, object labels are inherently given, so we are capable of extending the already existing UAV datasets with realistic quality images and high resolution meta-data. During the development of this thesis we have been able to produce a result of 68.4% on UAVid. This can be considered a new state-of-art result on this dataset.

Ecological Literacy And Beyond: Problem-based Learning For Future Professionals.

Ecological science contributes to solving a broad range of environmental problems. However, lack of ecological literacy in practice often limits application of this knowledge. In this paper, we highlight a critical but often overlooked demand on ecological literacy: to enable professionals of various careers to apply scientific knowledge when faced with environmental problems. Current university courses on ecology often fail to persuade students that ecological science provides important tools for environmental problem solving. We propose problem-based learning to improve the understanding of ecological science and its usefulness for real-world environmental issues that professionals in careers as diverse as engineering, public health, architecture, social sciences, or management will address. Courses should set clear learning objectives for cognitive skills they expect students to acquire. Thus, professionals in different fields will be enabled to improve environmental decision-making processes and to participate effectively in multidisciplinary work groups charged with tackling environmental issues.

Flicker perimetry in healthy subjects: influence of age and gender, learning effect and short-term fluctuation

PURPOSE: To determine the mean critical fusion frequency and the short-term fluctuation, to analyze the influence of age, gender, and the learning effect in healthy subjects undergoing flicker perimetry. METHODS: Study 1 - 95 healthy subjects underwent flicker perimetry once in one eye. Mean critical fusion frequency values were compared between genders, and the influence of age was evaluated using linear regression analysis. Study 2 - 20 healthy subjects underwent flicker perimetry 5 times in one eye. The first 3 sessions were separated by an interval of 1 to 30 days, whereas the last 3 sessions were performed within the same day. The first 3 sessions were used to investigate the presence of a learning effect, whereas the last 3 tests were used to calculate short-term fluctuation. RESULTS: Study 1 - Linear regression analysis demonstrated that mean global, foveal, central, and critical fusion frequency per quadrant significantly decreased with age (p<0.05).There were no statistically significant differences in mean critical fusion frequency values between males and females (p>0.05), with the exception of the central area and inferonasal quadrant (p=0.049 and p=0.011, respectively), where the values were lower in females. Study 2 - Mean global (p=0.014), central (p=0.008), and peripheral (p=0.03) critical fusion frequency were significantly lower in the first session compared to the second and third sessions. The mean global short-term fluctuation was 5.06±1.13 Hz, the mean interindividual and intraindividual variabilities were 11.2±2.8% and 6.4±1.5%, respectively. CONCLUSION: This study suggests that, in healthy subjects, critical fusion frequency decreases with age, that flicker perimetry is associated with a learning effect, and that a moderately high short-term fluctuation is expected.

Sensitivity and specificity of machine learning classifiers for glaucoma diagnosis using Spectral Domain OCT and standard automated perimetry

PURPOSE: To evaluate the sensitivity and specificity of machine learning classifiers (MLCs) for glaucoma diagnosis using Spectral Domain OCT (SD-OCT) and standard automated perimetry (SAP). METHODS: Observational cross-sectional study. Sixty two glaucoma patients and 48 healthy individuals were included. All patients underwent a complete ophthalmologic examination, achromatic standard automated perimetry (SAP) and retinal nerve fiber layer (RNFL) imaging with SD-OCT (Cirrus HD-OCT; Carl Zeiss Meditec Inc., Dublin, California). Receiver operating characteristic (ROC) curves were obtained for all SD-OCT parameters and global indices of SAP. Subsequently, the following MLCs were tested using parameters from the SD-OCT and SAP: Bagging (BAG), Naive-Bayes (NB), Multilayer Perceptron (MLP), Radial Basis Function (RBF), Random Forest (RAN), Ensemble Selection (ENS), Classification Tree (CTREE), Ada Boost M1(ADA),Support Vector Machine Linear (SVML) and Support Vector Machine Gaussian (SVMG). Areas under the receiver operating characteristic curves (aROC) obtained for isolated SAP and OCT parameters were compared with MLCs using OCT+SAP data. RESULTS: Combining OCT and SAP data, MLCs' aROCs varied from 0.777(CTREE) to 0.946 (RAN).The best OCT+SAP aROC obtained with RAN (0.946) was significantly larger the best single OCT parameter (p<0.05), but was not significantly different from the aROC obtained with the best single SAP parameter (p=0.19). CONCLUSION: Machine learning classifiers trained on OCT and SAP data can successfully discriminate between healthy and glaucomatous eyes. The combination of OCT and SAP measurements improved the diagnostic accuracy compared with OCT data alone.