Integrated visual perception architecture for robotic clothes perception and manipulation

This thesis proposes a generic visual perception architecture for robotic clothes perception and manipulation. This proposed architecture is fully integrated with a stereo vision system and a dual-arm robot and is able to perform a number of autonomous laundering tasks. Clothes perception and manipulation is a novel research topic in robotics and has experienced rapid development in recent years. Compared to the task of perceiving and manipulating rigid objects, clothes perception and manipulation poses a greater challenge. This can be attributed to two reasons: firstly, deformable clothing requires precise (high-acuity) visual perception and dexterous manipulation; secondly, as clothing approximates a non-rigid 2-manifold in 3-space, that can adopt a quasi-infinite configuration space, the potential variability in the appearance of clothing items makes them difficult to understand, identify uniquely, and interact with by machine. From an applications perspective, and as part of EU CloPeMa project, the integrated visual perception architecture refines a pre-existing clothing manipulation pipeline by completing pre-wash clothes (category) sorting (using single-shot or interactive perception for garment categorisation and manipulation) and post-wash dual-arm flattening. To the best of the author’s knowledge, as investigated in this thesis, the autonomous clothing perception and manipulation solutions presented here were first proposed and reported by the author. All of the reported robot demonstrations in this work follow a perception-manipulation method- ology where visual and tactile feedback (in the form of surface wrinkledness captured by the high accuracy depth sensor i.e. CloPeMa stereo head or the predictive confidence modelled by Gaussian Processing) serve as the halting criteria in the flattening and sorting tasks, respectively. From scientific perspective, the proposed visual perception architecture addresses the above challenges by parsing and grouping 3D clothing configurations hierarchically from low-level curvatures, through mid-level surface shape representations (providing topological descriptions and 3D texture representations), to high-level semantic structures and statistical descriptions. A range of visual features such as Shape Index, Surface Topologies Analysis and Local Binary Patterns have been adapted within this work to parse clothing surfaces and textures and several novel features have been devised, including B-Spline Patches with Locality-Constrained Linear coding, and Topology Spatial Distance to describe and quantify generic landmarks (wrinkles and folds). The essence of this proposed architecture comprises 3D generic surface parsing and interpretation, which is critical to underpinning a number of laundering tasks and has the potential to be extended to other rigid and non-rigid object perception and manipulation tasks. The experimental results presented in this thesis demonstrate that: firstly, the proposed grasp- ing approach achieves on-average 84.7% accuracy; secondly, the proposed flattening approach is able to flatten towels, t-shirts and pants (shorts) within 9 iterations on-average; thirdly, the proposed clothes recognition pipeline can recognise clothes categories from highly wrinkled configurations and advances the state-of-the-art by 36% in terms of classification accuracy, achieving an 83.2% true-positive classification rate when discriminating between five categories of clothes; finally the Gaussian Process based interactive perception approach exhibits a substantial improvement over single-shot perception. Accordingly, this thesis has advanced the state-of-the-art of robot clothes perception and manipulation.

Learning Binary Code Representations for Effective and Efficient Image Retrieval

The size of online image datasets is constantly increasing. Considering an image dataset with millions of images, image retrieval becomes a seemingly intractable problem for exhaustive similarity search algorithms. Hashing methods, which encodes high-dimensional descriptors into compact binary strings, have become very popular because of their high efficiency in search and storage capacity. In the first part, we propose a multimodal retrieval method based on latent feature models. The procedure consists of a nonparametric Bayesian framework for learning underlying semantically meaningful abstract features in a multimodal dataset, a probabilistic retrieval model that allows cross-modal queries and an extension model for relevance feedback. In the second part, we focus on supervised hashing with kernels. We describe a flexible hashing procedure that treats binary codes and pairwise semantic similarity as latent and observed variables, respectively, in a probabilistic model based on Gaussian processes for binary classification. We present a scalable inference algorithm with the sparse pseudo-input Gaussian process (SPGP) model and distributed computing. In the last part, we define an incremental hashing strategy for dynamic databases where new images are added to the databases frequently. The method is based on a two-stage classification framework using binary and multi-class SVMs. The proposed method also enforces balance in binary codes by an imbalance penalty to obtain higher quality binary codes. We learn hash functions by an efficient algorithm where the NP-hard problem of finding optimal binary codes is solved via cyclic coordinate descent and SVMs are trained in a parallelized incremental manner. For modifications like adding images from an unseen class, we propose an incremental procedure for effective and efficient updates to the previous hash functions. Experiments on three large-scale image datasets demonstrate that the incremental strategy is capable of efficiently updating hash functions to the same retrieval performance as hashing from scratch.

Efficient Image and Video Representations for Retrieval

Image (Video) retrieval is an interesting problem of retrieving images (videos) similar to the query. Images (Videos) are represented in an input (feature) space and similar images (videos) are obtained by finding nearest neighbors in the input representation space. Numerous input representations both in real valued and binary space have been proposed for conducting faster retrieval. In this thesis, we present techniques that obtain improved input representations for retrieval in both supervised and unsupervised settings for images and videos. Supervised retrieval is a well known problem of retrieving same class images of the query. We address the practical aspects of achieving faster retrieval with binary codes as input representations for the supervised setting in the first part, where binary codes are used as addresses into hash tables. In practice, using binary codes as addresses does not guarantee fast retrieval, as similar images are not mapped to the same binary code (address). We address this problem by presenting an efficient supervised hashing (binary encoding) method that aims to explicitly map all the images of the same class ideally to a unique binary code. We refer to the binary codes of the images as `Semantic Binary Codes' and the unique code for all same class images as `Class Binary Code'. We also propose a new class­ based Hamming metric that dramatically reduces the retrieval times for larger databases, where only hamming distance is computed to the class binary codes. We also propose a Deep semantic binary code model, by replacing the output layer of a popular convolutional Neural Network (AlexNet) with the class binary codes and show that the hashing functions learned in this way outperforms the state­ of ­the art, and at the same time provide fast retrieval times. In the second part, we also address the problem of supervised retrieval by taking into account the relationship between classes. For a given query image, we want to retrieve images that preserve the relative order i.e. we want to retrieve all same class images first and then, the related classes images before different class images. We learn such relationship aware binary codes by minimizing the similarity between inner product of the binary codes and the similarity between the classes. We calculate the similarity between classes using output embedding vectors, which are vector representations of classes. Our method deviates from the other supervised binary encoding schemes as it is the first to use output embeddings for learning hashing functions. We also introduce new performance metrics that take into account the related class retrieval results and show significant gains over the state­ of­ the art. High Dimensional descriptors like Fisher Vectors or Vector of Locally Aggregated Descriptors have shown to improve the performance of many computer vision applications including retrieval. In the third part, we will discuss an unsupervised technique for compressing high dimensional vectors into high dimensional binary codes, to reduce storage complexity. In this approach, we deviate from adopting traditional hyperplane hashing functions and instead learn hyperspherical hashing functions. The proposed method overcomes the computational challenges of directly applying the spherical hashing algorithm that is intractable for compressing high dimensional vectors. A practical hierarchical model that utilizes divide and conquer techniques using the Random Select and Adjust (RSA) procedure to compress such high dimensional vectors is presented. We show that our proposed high dimensional binary codes outperform the binary codes obtained using traditional hyperplane methods for higher compression ratios. In the last part of the thesis, we propose a retrieval based solution to the Zero shot event classification problem - a setting where no training videos are available for the event. To do this, we learn a generic set of concept detectors and represent both videos and query events in the concept space. We then compute similarity between the query event and the video in the concept space and videos similar to the query event are classified as the videos belonging to the event. We show that we significantly boost the performance using concept features from other modalities.

Dynamic mental representations of habitual behaviours: Food choice on a web-based environment

Aim: Rather than being rigid, habitual behaviours may be determined by dynamic mental representations that can adapt to context changes. This adaptive potential may result from particular conditions dependent on the interaction between two sources of mental constructs activation: perceived context applicability and cognitive accessibility . Method: T wo web-shopping simulations of fering the choice between habitually chosen and non-habitually chosen food products were presented to participants. This considered two choice contexts dif fering in the habitual behaviour perceived applicability (low vs. high) and a measure of habitual behaviour chronicity . Results: Study 1 demonstrated a perceived applicability ef fect, with more habitual (non-organic) than non-habitual (organic) food products chosen in a high perceived applicability (familiar) than in a low perceived applicability (new) context. The adaptive potential of habitual behaviour was evident in the habitual products choice consistency across three successive choices, despite the decrease in perceived applicability . Study 2 evidenced the adaptive potential in strong habitual behaviour participants – high chronic accessibility – who chose a habitual product (milk) more than a non-habitual product (orange juice), even when perceived applicability was reduced (new context). Conclusion: Results portray consumers as adaptive decision makers that can flexibly cope with changes in their (inner and outer) choice contexts.

Visual Insight in Geometry

According to a traditional rationalist proposal, it is possible to attain knowledge of certain necessary truths by means of insight—an epistemic mental act that combines the 'presentational' character of perception with the a priori status usually reserved for discursive reasoning. In this dissertation, I defend the insight proposal in relation to a specific subject matter: elementary Euclidean plane geometry, as set out in Book I of Euclid's Elements. In particular, I argue that visualizations and visual experiences of diagrams allow human subjects to grasp truths of geometry by means of visual insight. In the first two chapters, I provide an initial defense of the geometrical insight proposal, drawing on a novel interpretation of Plato's Meno to motivate the view and to reply to some objections. In the remaining three chapters, I provide an account of the psychological underpinnings of geometrical insight, a task that requires considering the psychology of visual imagery alongside the details of Euclid's geometrical system. One important challenge is to explain how basic features of human visual representations can serve to ground our intuitive grasp of Euclid's postulates and other initial assumptions. A second challenge is to explain how we are able to grasp general theorems by considering diagrams that depict only special cases. I argue that both of these challenges can be met by an account that regards geometrical insight as based in visual experiences involving the combined deployment of two varieties of 'dynamic' visual imagery: one that allows the subject to visually rehearse spatial transformations of a figure's parts, and another that allows the subject to entertain alternative ways of structurally integrating the figure as a whole. It is the interplay between these two forms of dynamic imagery that enables a visual experience of a diagram, suitably animated in visual imagination, to justify belief in the propositions of Euclid’s geometry. The upshot is a novel dynamic imagery account that explains how intuitive knowledge of elementary Euclidean plane geometry can be understood as grounded in visual insight.

Veteranness : Representations of Combat-related PTSD in U.S. Popular Visual Media

Posttraumatic stress and PTSD are becoming familiar terms to refer to what we often call the invisible wounds of war, yet these are recent additions to a popular discourse in which images of and ideas about combat-affected veterans have long circulated. A legacy of ideas about combat veterans and war trauma thus intersects with more recent clinical information about PTSD to become part of a discourse of visual media that has defined and continues to redefine veteran for popular audiences. In this dissertation I examine realist combat veteran representations in selected films and other visual media from three periods: during and after World Wars I and II (James Allen from I Am a Fugitive from a Chain Gang, Fred Derry and Al Stephenson from The Best Years of Our Lives); after the Vietnam War (Michael from The Deer Hunter, Eriksson from Casualties of War), and post 9/11 (Will James from The Hurt Locker, a collection of veterans from Wartorn: 1861-2010.) Employing a theoretical framework informed by visual media studies, Barthes’ concept of myth, and Foucault’s concept ofdiscursive unity, I analyze how these veteran representations are endowed with PTSD symptom-like behaviors and responses that seem reasonable and natural within the narrative arc. I contend that veteran myths appear through each veteran representation as the narrative develops and resolves. I argue that these veteran myths are many and varied but that they crystallize in a dominant veteran discourse, a discursive unity that I term veteranness. I further argue that veteranness entangles discrete categories such as veteran, combat veteran, and PTSD with veteran myths, often tying dominant discourse about combat-related PTSD to outdated or outmoded notions that significantly affect our attitudes about and treatment of veterans. A basic premise of my research is that unless and until we learn about the lasting effects of the trauma inherent to combat, we hinder our ability to fulfill our responsibilities to war veterans. A society that limits its understanding of posttraumatic stress, PTSD and post-war experiences of actual veterans affected by war trauma to veteranness or veteran myths risks normalizing or naturalizing an unexamined set of sociocultural expectations of all veterans, rendering them voice-less, invisible, and, ultimately disposable.

Towards flexible parsing of structured textual model representations

Existing parsers for textual model representation formats such as XMI and HUTN are unforgiving and fail upon even the smallest inconsistency between the structure and naming of metamodel elements and the contents of serialised models. In this paper, we demonstrate how a fuzzy parsing approach can transparently and automatically resolve a number of these inconsistencies, and how it can eventually turn XML into a human-readable and editable textual model representation format for particular classes of models.

Diseño y estudio de situaciones didácticas que favorecen el trabajo con registros semióticos

What’s behind the mistakes and difficulties that appear on the students to understand and study mathematics?are only related to the cognitive complexity of the content or such difficulties are also related to the possible ways to access the different mathematical objects? The mathematical activity generated in many students learning difficulties that are not manifested in cognitive processes related to other areas of knowledge. If something characterizes the processes of teaching and learning of mathematics is that, unlike what happens with the objects of study in the experimental sciences, the only way to access to them is through its different semiotic representations. The coordination among the different systems of representation that refer to the same mathematical concept, needs to move from one register to another (D’Amore, 1998, 2001, 2003, 2004, 2006; Duval, 1993, 1994, 1995, 1996, 2000, 2003, 2004, 2005, 2007, 2008, 2011, 2012; Godino, 2002, 2003, 2012, 2014; Kaput, 1989a, 1989b,1992, 1998; Radford, 1998, 2004a, 2004b, 2004c, 2006a, 2008,2009, 2011, 2013, 2014a). Therefore, the treatments that can be realized within a given register and the conversion of one register into another, play an essential role in the grasp of the object and mathematical concepts. Through this work with representations, students give meanings to the objects of study and are able to understand the underlying mathematical structures, which is the main educational interest of this issue...

Defending the 'public interest': an assessment of competing actor representations of 'solutions' to growing natural resource deficiencies.

This paper applies a SRT framework to the study of two case studies, namely the recent campaign of opposition to the legalization of hydraulic fracking in the State of New York and the more ongoing debate on land leasing in Africa. In relation to both campaigns, the analysis accounts for the arguments of a major financial institution and industry representatives who stress the safe and value-adding dimensions of these practices, as well as the views of opponents who refute the validity of industry's position and point to the unacceptable risks posed to the community, health and the environment. In spite of a number of obvious differences between these two case studies, not least differences arising from contrasting socio-economic and geo-political settings, there were also some notable similarities. First, was a tendency amongst protesters in both cases to formulate their role as contemporaries in a historically extended struggle for democratic justice. All perceived of themselves as guardians of their community's right to resist a corporate 'invasion' of their territories, like their forefathers and mothers before them. A theme of colonialism was explored in both settings through various identity and thematic anchoring devices that deliberately evoked shared understandings and historical memories of exploitation and human suffering. The evocation of powerful symbols of identity through visual narratives of protest further reinforced the cultural comprehensibility of opponents' message of protest in both contexts.

The potential of symbolic approximation. Disentangling the effects of approximation vs. calculation demands in nonsymbolic and symbolic representations.

Mathematical skills that we acquire during formal education mostly entail exact numerical processing. Besides this specifically human faculty, an additional system exists to represent and manipulate quantities in an approximate manner. We share this innate approximate number system (ANS) with other nonhuman animals and are able to use it to process large numerosities long before we can master the formal algorithms taught in school. Dehaene´s (1992) Triple Code Model (TCM) states that also after the onset of formal education, approximate processing is carried out in this analogue magnitude code no matter if the original problem was presented nonsymbolically or symbolically. Despite the wide acceptance of the model, most research only uses nonsymbolic tasks to assess ANS acuity. Due to this silent assumption that genuine approximation can only be tested with nonsymbolic presentations, up to now important implications in research domains of high practical relevance remain unclear, and existing potential is not fully exploited. For instance, it has been found that nonsymbolic approximation can predict math achievement one year later (Gilmore, McCarthy, & Spelke, 2010), that it is robust against the detrimental influence of learners´ socioeconomic status (SES), and that it is suited to foster performance in exact arithmetic in the short-term (Hyde, Khanum, & Spelke, 2014). We provided evidence that symbolic approximation might be equally and in some cases even better suited to generate predictions and foster more formal math skills independently of SES. In two longitudinal studies, we realized exact and approximate arithmetic tasks in both a nonsymbolic and a symbolic format. With first graders, we demonstrated that performance in symbolic approximation at the beginning of term was the only measure consistently not varying according to children´s SES, and among both approximate tasks it was the better predictor for math achievement at the end of first grade. In part, the strong connection seems to come about from mediation through ordinal skills. In two further experiments, we tested the suitability of both approximation formats to induce an arithmetic principle in elementary school children. We found that symbolic approximation was equally effective in making children exploit the additive law of commutativity in a subsequent formal task as a direct instruction. Nonsymbolic approximation on the other hand had no beneficial effect. The positive influence of the symbolic approximate induction was strongest in children just starting school and decreased with age. However, even third graders still profited from the induction. The results show that also symbolic problems can be processed as genuine approximation, but that beyond that they have their own specific value with regard to didactic-educational concerns. Our findings furthermore demonstrate that the two often con-founded factors ꞌformatꞌ and ꞌdemanded accuracyꞌ cannot be disentangled easily in first graders numerical understanding, but that children´s SES also influences existing interrelations between the different abilities tested here.

ESL teacher profiles of ICT integration in their classroom practices and assessment activities : a portrait viewed through the lens of some Quebec teachers’ social representations

Abstract : Information and communication technologies (ICTs, henceforth) have become ubiquitous in our society. The plethora of devices competing with the computer, from iPads to the Interactive whiteboard, just to name a few, has provided teachers and students alike with the ability to communicate and access information with unprecedented accessibility and speed. It is only logical that schools reflect these changes given that their purpose is to prepare students for the future. Surprisingly enough, research indicates that ICT integration into teaching activities is still marginal. Many elementary and secondary schoolteachers are not making effective use of ICTs in their teaching activities as well as in their assessment practices. The purpose of the current study is a) to describe Quebec ESL teachers’ profiles of using ICTs in their daily teaching activities; b) to describe teachers’ ICT integration and assessment practices; and c) to describe teachers’ social representations regarding the utility and relevance of ICT use in their daily teaching activities and assessment practices. In order to attain our objectives, we based our theoretical framework, principally, on the social representations (SR, henceforth) theory and we defined most related constructs which were deemed fundamental to the current thesis. We also collected data from 28 ESL elementary and secondary school teachers working in public and private sectors. The interview guide used to that end included a range of items to elicit teachers’ SR in terms of ICT daily use in teaching activities as well as in assessment practices. In addition, we carried out our data analyses from a textual statistics perspective, a particular mode of content analysis, in order to extract the indicators underlying teachers’ representations of the teachers. The findings suggest that although almost all participants use a wide range of ICT tools in their practices, ICT implementation is seemingly not exploited to its fullest potential and, correspondingly, is likely to produce limited effects on students’ learning. Moreover, none of the interviewees claim that they use ICTs in their assessment practices and they still hold to the traditional paper-based assessment (PBA, henceforth) approach of assessing students’ learning. Teachers’ common discourse reveals a gap between the positive standpoint with regards to ICT integration, on the one hand, and the actual uses of instructional technology, on the other. These results are useful for better understanding the way ESL teachers in Quebec currently view their use of ICTs, particularly for evaluation purposes. In fact, they provide a starting place for reconsidering the implementation of ICTs in elementary and secondary schools. They may also be useful to open up avenues for the development of a future research program in this regard.

Social representations about the Portuguese adolescent alcohol consumption

Alcohol is currently the most widely consumed psychoactive substance in the world and Portugal is the second country where such consumption is greater, registering a large increase in consumption by young people. Currently continue still, beliefs, myths and prejudices that because they are well rooted culturally serve as good reasons for drinking. This study sought therefore to identify the myths associated by adolescents to alcohol consumption. A questionnaire was developed for this purpose (74 items, α = 0.947) and applied to a sample of 1176 adolescents schooled between 14 and 18 years old, with a return rate of 42.6% (margin of error of 5% for a confidence level of 95%) in the district of Beja, Portugal, in 2012. The collected data were statistically analyzed using measures of association, factor analysis and linear regression. The results show that many myths are unknown among adolescents, verifying the presence of many questions, among which stands out: alcohol "warm", "thirst quenching", "gives strength", "facilitates digestion" "whet the appetite", "is a medicine", "is aphrodisiac", "facilitates social relations", among others. Age and sex are variables significantly affected the myths and objectives of alcohol consumption. These results clearly point to the need to be disassembled beliefs and wrong conceptions about the effects of alcohol consumption, particularly in the school environment, reducing the risk of the consequences and promoting adolescent health, preventing any future dependence on this psychoactive substance.

Social Representations of Violence against the Elderly

The repercussions of violence on the mental, social, and physical well-being of the elderly are some of the most challenging problems in public health today. Using a qualitative design, we conducted a study in Portugal and the United States that applied both descriptive and comparative methods in order to understand the social representations of violence against the elderly. Utilizing the Theory of Social Representations, we explored the perspectives of the elderly, their families, and healthcare professionals on the subject of violence against the elderly. The data on which the findings were based were obtained in two very different cultural contexts, yet the representations of violence against the elderly revealed no significant cross-cultural differences. However, conceptualizations regarding expectations of care and protection for the elderly proved to be distinct. We discussed concerns about the general attitudes of tolerance toward violence, including those of the elderly who self-identified as eventual victims. Violence against the elderly was portrayed as a part of old age and also somehow was justified by it. The results also indicated the need to better prepare healthcare professionals and society in general to deal with the consequences of the problem and not, as we would like to report, to prevent it from happening.