Fluent Human–Robot Dialogues About Grounded Objects in Home Environments

Persson, Andreas; Moubayed, Samer Al; Loutfi, Amy

doi:10.1007/s12559-014-9291-y

Cited by 4 publications

(4 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed algorithm was implemented on top of the OpenCV library 1 . The oriented FAST keypoint detector [6] was used as a keypoint detector.…”

Section: Discussionmentioning

confidence: 99%

“…However, an improvement in computational efficiency is often made at the cost of a reduced retrieval accuracy (and vice versa) and a general trend is therefore to use computationally efficient approximated solutions supported by additional learning (either supervised or unsupervised) via a demanding off-line training phase. Yet while off-line training is acceptable for creating a reference space of objects and scenes for recognition tasks, a real time system such as an operating robot platform must also maintain a dynamic representation of recognized objects and/or scenes [1]. For this purpose, the representation of features must not only be compact but also dynamic, so that the algorithm can adapt and incorporate new features without repeating elaborate off-line training.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Fast Matching of Binary Descriptors for Large-Scale Applications in Robot Vision

Persson

Loutfi

2016

International Journal of Advanced Robotic Systems

Self Cite

View full text Add to dashboard Cite

The introduction of computationally efficient binary feature descriptors has raised new opportunities for realworld robot vision applications. However, brute force feature matching of binary descriptors is only practical for smaller datasets. In the literature, there has therefore been an increasing interest in representing and matching binary descriptors more efficiently. In this article, we follow this trend and present a method for efficiently and dynamically quantizing binary descriptors through a summarized frequency count into compact representations (called fsum) for improved feature matching of binary pointfeatures. With the motivation that real-world robot applications must adapt to a changing environment, we further present an overview of the field of algorithms, which concerns the efficient matching of binary descriptors and which are able to incorporate changes over time, such as clustered search trees and bag-of-features improved by vocabulary adaptation. The focus for this article is on evaluation, particularly large scale evaluation, compared to alternatives that exist within the field. Throughout this evaluation it is shown that the fsum approach is both efficient in terms of computational cost and memory requirements, while retaining adequate retrieval accuracy. It is further shown that the presented algorithm is equally suited to binary descriptors of arbitrary type and that the algorithm is therefore a valid option for several types of vision applications.

show abstract

“…The proposed algorithm was implemented on top of the OpenCV library 1 . The oriented FAST keypoint detector [6] was used as a keypoint detector.…”

Section: Discussionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Fast Matching of Binary Descriptors for Large-Scale Applications in Robot Vision

Persson

Loutfi

2016

International Journal of Advanced Robotic Systems

Self Cite

View full text Add to dashboard Cite

show abstract

“…Different from conventional computing technologies, cognitive computing aims at addressing complex, ambiguous and uncertain problems by mimicking the cognitive process of human [29,30,37]. In order to achieve this new level of computing, cognitive systems have to be adaptive, in the sense that they must learn from changing data, resolve ambiguity and tolerate unpredictability of the environment in real time, or near real time [10,16,19].…”

Section: Introductionmentioning

confidence: 99%

Self-Adaptive Windowing Approach for Handling Complex Concept Drift

et al. 2015

View full text Add to dashboard Cite

Detecting changes in data streams attracts major attention in cognitive computing systems. The challenging issue is how to monitor and detect these changes in order to preserve the model performance during complex drifts. By complex drift, we mean a drift that presents many characteristics in the sometime. The most challenging complex drifts are gradual continuous drifts, where changes are only noticed during a long time period. Moreover, these gradual drifts may also be local, in the sense that they may affect a little amount of data, and thus make the drift detection more complicated. For this purpose, a new drift detection mechanism, EDIST2, is proposed in order to deal with these complex drifts. EDIST2 monitors the learner performance through a self-adaptive window that is autonomously adjusted through a statistical hypothesis test. This statistical test provides theoretical guarantees, regarding the false alarm rate, which were experimentally confirmed. EDIST2 has been tested through six synthetic datasets presenting different kinds of complex drift, and five real-world datasets. Encouraging results were found, comparing to similar approaches, where EDIST2 has achieved good accuracy rate in synthetic and realworld datasets and has achieved minimum delay of detection and false alarm rate.

show abstract

“…In this context, the appropriate design and ability of both a VUI and DMS in processing and understanding free forms of conversations allow users to feel a naturalistic interaction and ensure their satisfaction [14,15]. To our knowledge, there are no standards for the development of such satisfying DMSs and VUIs, and the contributions of Marcin Skowron et al [21], Ingo Siegert et al [20], Andreas Persson et al [17] and Jiří Přibil and Anna Přibilová [18] are all providing original solutions for improving quality standards of these devices.…”

mentioning

confidence: 99%

Modeling Emotion, Behavior and Context in Socially Believable Robots and ICT Interfaces

Esposito

Fortunati

Lugano³

2014

Cogn Comput

View full text Add to dashboard Cite

The modeling and implementation of sophisticated multimodal software/hardware interfaces is a current scientific challenge of high societal relevance. The main characteristics entailed by these interfaces are being able to interact with people, inferring social, organizational and physical contexts based on sensed data, assisting people with special needs, enhancing elderly health-care assistance, learning and rehabilitation in daily functional activities. Implementing such Human Computer Interaction (HCI) systems is of public utility and profitable for a living science that should simplify user's accesses to a wide range of social services, either remotely or in a person-to-person setting.The current and future applications foreseen in this highly interdisciplinary field are countless: among these are featured context-aware avatars and robotic devices replacing and/or acting on behalf of humans in high responsibility tasks or time-critical dangerous tasks such as urban emergencies. Other emerging applications concern robot companions for elderly and vulnerable people and intelligent agents for services where there is a shortage of suitable skills or otherwise there is a request of significant investments in training-qualified personnel such as in therapist-based interventions. Given the complexities required by these automated tasks, the approach for developing such devices has to account for a holistic investigation perspective. New cognitive architectures must be foreseen and new cognitive integrations must be exploited in order to take advantage of the knowledge derived from the analysis of human behaviors across different contexts.At the stake, there is the need to develop a deep understanding of the emotional and intentional cognitive processes underpinning human interactions. Inherently new insights must be deployed for designing complex-autonomous systems, which are required to be able to feel human emotional and intentional states; cooperatively adapt to them through a socially ethical and sensible conduct; and exhibit coherent vocal, visual and gestural affordances.The present Special Issue investigates these topics, by gathering new experimental data and theories across a spectrum of disciplines, in order to identify the metastructures underlying these phenomena. This effort hopefully will stimulate, on the one hand, the conception of new mathematical models for representing data, reasoning and learning. On the other hand, it will produce new psychological and computational approaches with respect to the existing cognitive frameworks and algorithmic solutions.Enabling a consistent progress toward the implementation of a human automaton level of intelligence is crucial for developing such HCI systems and enhancing the quality of life of people addressing their current and future societal needs.The topics proposed by the present special issue are interdisciplinary and cover issues related to several areas of research. Let us report them: behavioral analysis of interactions; mathematical models for repre...

show abstract

Fluent Human–Robot Dialogues About Grounded Objects in Home Environments

Cited by 4 publications

References 27 publications

Fast Matching of Binary Descriptors for Large-Scale Applications in Robot Vision

Fast Matching of Binary Descriptors for Large-Scale Applications in Robot Vision

Self-Adaptive Windowing Approach for Handling Complex Concept Drift

Modeling Emotion, Behavior and Context in Socially Believable Robots and ICT Interfaces

Contact Info

Product

Resources

About