oru.sePublications
Change search
Link to record
Permanent link

Direct link
BETA
Bhatt, Mehul, ProfessorORCID iD iconorcid.org/0000-0002-6290-5492
Biography [eng]

 

 
Biography [swe]

 

 
Publications (10 of 99) Show all publications
Bhatt, M., Suchan, J. & Vardarajan, S. (2019). Deep Semantics for Explainable Visuospatial Intelligence: Perspectives on Integrating Commonsense Spatial Abstractions and Low-Level Neural Features. In: Derek Doran; Artur d'Avila Garcez; Freddy Lecue (Ed.), Proceedings of the 2019 International Workshop on Neural-Symbolic Learning and Reasoning: Annual workshop of the Neural-Symbolic Learning and Reasoning Association. Paper presented at 14th International Workshop on Neural-Symbolic Learning and Reasoning (NeSy’19) in conjunction with IJCAI 2019, Macao, China, August 10-16, 2019.
Open this publication in new window or tab >>Deep Semantics for Explainable Visuospatial Intelligence: Perspectives on Integrating Commonsense Spatial Abstractions and Low-Level Neural Features
2019 (English)In: Proceedings of the 2019 International Workshop on Neural-Symbolic Learning and Reasoning: Annual workshop of the Neural-Symbolic Learning and Reasoning Association / [ed] Derek Doran; Artur d'Avila Garcez; Freddy Lecue, 2019Conference paper, Published paper (Refereed)
Abstract [en]

High-level semantic interpretation of (dynamic) visual imagery calls for general and systematic methods integrating techniques in knowledge representation and computer vision. Towards this, we position "deep semantics", denoting the existence of declarative models –e.g., pertaining "space and motion"– and corresponding formalisation and methods supporting (domain-independent) explainability capabilities such as semantic question-answering, relational (and relationally-driven) visuospatial learning, and (non-monotonic) visuospatial abduction. Rooted in recent work, we summarise and report the status quo on deep visuospatial semantics —and our approach to neurosymbolic integration and explainable visuo-spatial computing in that context— with developed methods and tools in diverse settings such as behavioural research in psychology, art & social sciences, and autonomous driving.

National Category
Computer Vision and Robotics (Autonomous Systems) Media and Communication Technology Computer Sciences
Research subject
Computer Science
Identifiers
urn:nbn:se:oru:diva-76859 (URN)
Conference
14th International Workshop on Neural-Symbolic Learning and Reasoning (NeSy’19) in conjunction with IJCAI 2019, Macao, China, August 10-16, 2019
Available from: 2019-10-01 Created: 2019-10-01 Last updated: 2019-10-07Bibliographically approved
Suchan, J., Bhatt, M. & Varadarajan, S. (2019). Out of Sight But Not Out of Mind: An Answer Set Programming Based Online Abduction Framework for Visual Sensemaking in Autonomous Driving. In: Sarit Kraus (Ed.), Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI-19): . Paper presented at 28th International Joint Conference on Artificial Intelligence (IJCAI 2019), Macao, China, August 10-16, 2019 (pp. 1879-1885). ijcai.org
Open this publication in new window or tab >>Out of Sight But Not Out of Mind: An Answer Set Programming Based Online Abduction Framework for Visual Sensemaking in Autonomous Driving
2019 (English)In: Proceedings of the Twenty-Eighth International Joint Conference on Artificial Intelligence (IJCAI-19) / [ed] Sarit Kraus, ijcai.org , 2019, p. 1879-1885Conference paper, Published paper (Refereed)
Abstract [en]

We demonstrate the need and potential of systematically integrated vision and semantics solutions for visual sensemaking (in the backdrop of autonomous driving). A general method for online visual sensemaking using answer set programming is systematically formalised and fully implemented. The method integrates state of the art in (deep learning based) visual computing, and is developed as a modular framework usable within hybrid architectures for perception & control. We evaluate and demo with community established benchmarks KITTIMOD and MOT. As use-case, we focus on the significance of human-centred visual sensemaking ---e.g., semantic representation and explainability, question-answering, commonsense interpolation--- in safety-critical autonomous driving situations.

Place, publisher, year, edition, pages
ijcai.org, 2019
National Category
Computer Sciences Computer Vision and Robotics (Autonomous Systems) Human Computer Interaction
Research subject
Computer Science; Automatic Control
Identifiers
urn:nbn:se:oru:diva-76857 (URN)10.24963/ijcai.2019/260 (DOI)978-0-9992411-4-1 (ISBN)
Conference
28th International Joint Conference on Artificial Intelligence (IJCAI 2019), Macao, China, August 10-16, 2019
Note

(Distinguished Paper Nomination - Received honourable Mention)

Available from: 2019-10-01 Created: 2019-10-01 Last updated: 2019-10-01Bibliographically approved
Li, B., Bhatt, M. & Schultz, C. P. L. (2019). λProlog(QS): Functional spatial reasoning in higher order logic programming. In: : . Paper presented at 14th International Conference on Spatial Information Theory (COSIT 2019), Regensburg, Germany, September 9-13, 2019. Schloss Dagstuhl - Leibniz-Zentrum für Informatik, 142, Article ID 26.
Open this publication in new window or tab >>λProlog(QS): Functional spatial reasoning in higher order logic programming
2019 (English)Conference paper, Published paper (Refereed)
Abstract [en]

We present a framework and proof-of-concept implementation for functional spatial reasoning within high-order logic programming. The developed approach extends λProlog to support reasoning over spatial variables via Constraint Handling Rules. We implement our approach within Embeddable λProlog Interpreter (ELPI) and demonstrate key features from combined reasoning over spatial functions and relations. The reported research is an ongoing development of the declarative spatial reasoning paradigm.

Place, publisher, year, edition, pages
Schloss Dagstuhl - Leibniz-Zentrum für Informatik, 2019
Series
Leibniz International Proceedings in Informatics, LIPIcs, E-ISSN 1868-8969
Keywords
Spatial reasoning, Functional logic programming, Lambda-Prolog
National Category
Computer Sciences
Research subject
Computer Science
Identifiers
urn:nbn:se:oru:diva-76858 (URN)10.4230/LIPIcs.COSIT.2019.26 (DOI)2-s2.0-85072631327 (Scopus ID)
Conference
14th International Conference on Spatial Information Theory (COSIT 2019), Regensburg, Germany, September 9-13, 2019
Note

Funding Agency:

Danish Independent Research Fund

Available from: 2019-10-01 Created: 2019-10-01 Last updated: 2019-10-07Bibliographically approved
Schultz, C. P. L., Bhatt, M., Suchan, J. & Walega, P. A. (2018). Answer Set Programming Modulo ’Space-Time’. In: Christoph Benzmüller; Francesco Ricca; Xavier Parent; Dumitru Roman (Ed.), Proceedings of the International Joint Conference on Rules and Reasoning: . Paper presented at 2nd International Joint Conference on Rules and Reasoning (RuleML+RR), Luxembourg, Luxembourg, September 18-21, 2018 (pp. 318-326). Springer, 11092
Open this publication in new window or tab >>Answer Set Programming Modulo ’Space-Time’
2018 (English)In: Proceedings of the International Joint Conference on Rules and Reasoning / [ed] Christoph Benzmüller; Francesco Ricca; Xavier Parent; Dumitru Roman, Springer, 2018, Vol. 11092, p. 318-326Conference paper, Published paper (Refereed)
Abstract [en]

We present ASP Modulo ‘Space-Time’, a declarative representational and computational framework to perform commonsense reasoning about regions with both spatial and temporal components. Supported are capabilities for mixed qualitative-quantitative reasoning, consistency checking, and inferring compositions of space-time relations; these capabilities combine and synergise for applications in a range of AI application areas where the processing and interpretation of spatio-temporal data is crucial. The framework and resulting system is the only general KR-based method for declaratively reasoning about the dynamics of ‘space-time’ regions as first-class objects.

Place, publisher, year, edition, pages
Springer, 2018
Series
Lecture Notes in Computer Science, ISSN 0302-9743, E-ISSN 1611-3349 ; 11092
National Category
Computer Vision and Robotics (Autonomous Systems) Computer Sciences
Research subject
Computer Science
Identifiers
urn:nbn:se:oru:diva-69933 (URN)10.1007/978-3-319-99906-7_24 (DOI)2-s2.0-85053202559 (Scopus ID)
Conference
2nd International Joint Conference on Rules and Reasoning (RuleML+RR), Luxembourg, Luxembourg, September 18-21, 2018
Available from: 2018-10-29 Created: 2018-10-29 Last updated: 2018-10-31Bibliographically approved
Bhatt, M. (2018). Cognitive media studies: Potentials for spatial cognition and AI research. Cognitive Processing, 19(Suppl. 1), S6-S6
Open this publication in new window or tab >>Cognitive media studies: Potentials for spatial cognition and AI research
2018 (English)In: Cognitive Processing, ISSN 1612-4782, E-ISSN 1612-4790, Vol. 19, no Suppl. 1, p. S6-S6Article in journal, Meeting abstract (Other academic) Published
Abstract [en]

Cognitive media studies has developed as an area of research at the interface of disciplines as diverse as aesthetics, psychology, neuroscience, film theory, and cognitive science. In this context, the focus of this talk is on the foundational significance of artificial intelligence and visuo-spatial cognition and computation for the design of inte-grated analytical–empirical methods for the (multi-modal) analysis of human behaviour data vis-a-vis a range of digital visuo-auditory narrative media (e.g., narrative film). The presentation focusses on the methodological foundations and assistive technologies for systematic formalization and empirical analyses aimed at, for instance, the generation of evidence, establishing and characterizing correlates between principles for the synthesis of the moving image (e.g., from a cinematographic viewpoint), and its perceptual recipient effects and influence on observers.

In the backdrop a range of completed and ongoing experiments, we emphasize the core results on the semantic interpretation of human behaviour vis-a-vis narrative film and its visuo-auditory reception. We demonstrate the manner in which AI-based models for machine coding of narrative, and relational inference and learning serves as basis to externalize explicit and inferred knowledge about embodied visuo-auditory reception, e.g., using modalities such as diagrammatic representations, natural language, complex (dynamic) data visualizations.

Demonstration: The presentation will particularly showcase methods and tools developed to perform perceptual narrativisation or sensemaking with multi-modal, dynamic human-behaviour data (combining visuo-spatial imagery such as film/video, eye-tracking, head-tracking during a perception task) for a chosen set of experimental material based on existing films, as well as lab-developed experimental content.

Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2018
National Category
Psychology
Identifiers
urn:nbn:se:oru:diva-68757 (URN)10.1007/s10339-018-0884-3 (DOI)000442849900015 ()
Available from: 2018-09-10 Created: 2018-09-10 Last updated: 2018-09-10Bibliographically approved
Kondyli, V. & Bhatt, M. (2018). Decision points in architectural space: How they affect users' visuo-locomotive experience during wayfinding. Cognitive Processing, 19(Suppl. 1), S43-S43
Open this publication in new window or tab >>Decision points in architectural space: How they affect users' visuo-locomotive experience during wayfinding
2018 (English)In: Cognitive Processing, ISSN 1612-4782, E-ISSN 1612-4790, Vol. 19, no Suppl. 1, p. S43-S43Article in journal, Meeting abstract (Other academic) Published
Abstract [en]

Decision points in a wayfinding path are considered not only the intersections but also changes in geometry and in directions, merging of paths, or transitions. Carpman and Simmon (1986) pinpoint the need for environmental cues in these points where users’ confusion arises. In this study, we investigate the morphology and the manifest cues of the decision points in relation to the visuo-locomotive behaviour of users recorded during a wayfinding case-study conducted in two healthcare buildings at the Parkland Hospital (Dallas).

We collect and analyse the embodied visuo-locomotive experience of 25 participants, using eye-tracking, external cameras, behavioural mapping, questionnaires, interviews, and orientations tasks. In our multi-modal qualitative analysis, founded in Spatial Reasoning, Cognitive Vision, and Environmental Psychology, we focus on the aspects of visual perception, decision making, orientation, and spatial knowledge acquisition. The comparison between users’ transition in eight decision points involves correlations between occurrences of confusion-related events, detection and categorisation of manifest cues, navigation performance, as well as visual attention analysis in relation to the available spatial features.

Primary results suggest that (1) stop and looking-around behaviour mostly emerge in the decision points; (2) behaviour that indicates confusion is mostly encoded in narrow and enclosed decision points; (3) transitional spaces intensify visual search; (4) visibility ahead of time, and visual disruptions affect the visuo-locomotive behaviour; and (5) detection of manifest cues is affected by the morphology of decision points. The correlations between behavioural and morphological data encoded to conceptual language can be useful as a baseline for computationally-driven behavioural analysis.

Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2018
National Category
Psychology
Identifiers
urn:nbn:se:oru:diva-68758 (URN)10.1007/s10339-018-0884-3 (DOI)000442849900135 ()
Available from: 2018-09-10 Created: 2018-09-10 Last updated: 2019-03-26Bibliographically approved
Bhatt, M. (2018). Embodied architecture design: On people-centered design of visuo-locomotive cognitive experiences. Cognitive Processing, 19(Suppl. 1), S5-S5
Open this publication in new window or tab >>Embodied architecture design: On people-centered design of visuo-locomotive cognitive experiences
2018 (English)In: Cognitive Processing, ISSN 1612-4782, E-ISSN 1612-4790, Vol. 19, no Suppl. 1, p. S5-S5Article in journal, Meeting abstract (Other academic) Published
Abstract [en]

This presentation focusses on the analysis and design of human-centered, embodied, cognitive user experiences from the perspectives of spatial cognition and computation, artificial intelligence, and human-computer interaction research. Focusing on large-scale built up spaces (in particular hospitals), this presentation will particularly address:

‘how can human-centered cognitive modalities of visuo-locomotive perception constitute the foundational building blocks of design education, discourse, systems, and the professional practice of spatial design for architecture’.

The presentation will emphasizeevidence-based multimodality studies from the viewpoints of visuo-locomotive (i.e., pertaining to vision, movement, and wayfinding) cognitive experiences. Modalities being investigated include: (1) visual attention (by eye-tracking), gesture, language, facial expressions; (2) human expert guided event segmentation (e.g., coming from behavioral or environmental psychologists, designers, annotators); (3) deep analysis based on dialogic components, think-aloud protocols. We demonstrate (1–3) in the context of a large-scale study conducted at the Old and New Parkland Hospitals in Dallas, Texas.

This research (and symposium) calls for a tightly integrated approach combining analytical methods (rooted in AI and computational cognition) and empirical methods (rooted in psychology and perception studies) for developing human-centered architectural design technologies, and technology-mediated (architectural) design synthesis.

Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2018
National Category
Psychology
Identifiers
urn:nbn:se:oru:diva-68756 (URN)10.1007/s10339-018-0884-3 (DOI)000442849900011 ()
Available from: 2018-09-10 Created: 2018-09-10 Last updated: 2018-09-10Bibliographically approved
Bhatt, M. (2018). Minds. Movement. Moving image. Cognitive Processing, 19(Suppl. 1), S5-S5
Open this publication in new window or tab >>Minds. Movement. Moving image
2018 (English)In: Cognitive Processing, ISSN 1612-4782, E-ISSN 1612-4790, Vol. 19, no Suppl. 1, p. S5-S5Article in journal, Meeting abstract (Other academic) Published
Abstract [en]

This symposium—conducted in two parts—explores the confluence of empirically-based qualitative research in the cognitive and psychological sciences (focusing on visual and spatial cognition) with computationally-driven analytical methods (rooted in artificial intelligence) in the service of communications, media, design, and human behavioural studies. With a focus on architecture and visuo-auditory media design, the twin-symposia will demonstrate recent results and explore the synergy of research methods for the study of human behaviour in the chosen (design) contexts of socio-cultural, and socio-technological significance.

The symposium brings together experts and addresses methodsand perspectives from:

•  Visuo-Spatial Cognition and Computation

•  Artificial Intelligence, Cognitive Systems

•  Multimodality and Interaction

•  Cognitive Science and Psychology

•  Neuroscience

•  Design Cognition and Computation

•  Communications and Media Studies

•  Architecture, Built Environment

•  Design Studies (focus on architecture and visuo-auditory media)

•  Evidence Based Design

The symposium particularly emphasises the role of multimodality and mediated interaction for the analysis and design of human-centered, embodied, cognitive user experiences in everyday life and work. Here, the focus is on multimodality studies aimed at the semantic interpretation of human behaviour, and the empirically-driven synthesis of embodied interactive experiences in real world settings. In focus are narrative media design, architecture and built environment design, product design, cognitive media studies (film, animation, VR, sound and music design), and user interaction studies. In these contexts, the symposium emphasizes evidence-based multimodality studies from the viewpoints of visual (e.g.,attention and recipient effects), visuo-locomotive (e.g. , movement, wayfinding), and visuo-auditory (e.g., narrative media) cognitive experiences. Modalities being investigated include, but are not limited to:

•  visual attention (by eye-tracking), gesture, speech, language, facial expressions, tactile interactions, olfaction, biosignals;

•  human expert guided event segmentation (e.g. coming from behavioral or environmental psychologists, designers, annotators,crowd-sensing)

•  deep analysis based on dialogic components, think-aloud protocols

The scientific agenda of the twin-symposia also emphasizes the multi-modality of the embodied visuo-spatial thinking involved in ‘‘problem-solving’’ for the design of objects, artefacts, and inter-active people-experiences emanating there from. Universality andinclusion in ‘‘design thinking’’ are of overarching focus in all design contexts relevant to the symposium; here, the implications of mul-timodality studies for inclusive design, e.g.,creation of presentations of the same content in different modalities, are also of interest. The symposium provides a platform to discuss the development of next-generation embodied interaction design systems, practices, and (human-centered) assistive frameworks and technologies encompassing the multi-faceted nature of embodied design conception and synthesis. Individual contributions/talks within the two symposia address the themes under consideration from formal, computational, cognitive, design, engineering, empirical, and philosophical perspectives.

Place, publisher, year, edition, pages
Springer Berlin/Heidelberg, 2018
National Category
Psychology
Identifiers
urn:nbn:se:oru:diva-68755 (URN)10.1007/s10339-018-0884-3 (DOI)000442849900010 ()
Available from: 2018-09-10 Created: 2018-09-10 Last updated: 2018-09-10Bibliographically approved
Kondyli, V., Bhatt, M. & Hartmann, T. (2018). Precedent Based Design Foundations for Parametric Design: The Case of Navigation and Wayfinding. Advances in Computational Design, 3(4), 339-366
Open this publication in new window or tab >>Precedent Based Design Foundations for Parametric Design: The Case of Navigation and Wayfinding
2018 (English)In: Advances in Computational Design, ISSN 2383-8477, Vol. 3, no 4, p. 339-366Article in journal (Refereed) Published
Abstract [en]

Parametric design systems serve as powerful assistive tools in the design process by providing a flexible approach for the generation of a vast number of design alternatives. However, contemporary parametric design systems focus primarily on low-level engineering and structural forms, without an explicit means to also take into account high-level, cognitively motivated people-centred design goals.

We present a precedent-based parametric design method that integrates people-centred design “precedents” rooted in empirical evidence directly within state of the art parametric design systems. As a use-case, we illustrate the general method in the context of an empirical study focusing on the multi-modal analysis of wayfinding behaviour in two large-scale healthcare environments. With this use-case, we demonstrate the manner in which: (1). a range of empirically established design precedents —e.g., pertaining to visibility and navigation— may be articulated as design constraints to be embedded directly within state of the art parametric design tools (e.g., Grasshopper); and (2). embedded design precedents lead to the (parametric) generation of a number of morphologies that satisfy people-centred design criteria (in this case, pertaining to wayfinding).

Our research presents an exemplar for the integration of cognitively motivated design goals with parametric design-space exploration methods. We posit that this opens-up a range of technological challenges for the engineering and development of next-generation computer aided architecture design systems.

Place, publisher, year, edition, pages
Techno-Press, 2018
Keywords
human behaviour studies, navigation, wayfinding, architecture design, spatial cognition, visual perception, parametric design, architectural computing, design computing
National Category
Computer and Information Sciences Human Aspects of ICT Architectural Engineering
Research subject
Computer Science; Psychology
Identifiers
urn:nbn:se:oru:diva-69940 (URN)10.12989/acd.2018.3.4.339 (DOI)000448366300002 ()
Available from: 2018-10-29 Created: 2018-10-29 Last updated: 2019-03-26Bibliographically approved
Bhatt, M. & Lieto, A. (Eds.). (2018). Proceedings of the 1st International Workshop on Cognition and Artificial Intelligence for Human-Centred Design 2017 co-located with IJCAI 2017. Paper presented at 1st International Workshop on Cognition and Artificial Intelligence for Human-Centred Design (CAID), Melbourne, Australia, August 19, 2017. Technical University of Aachen, 2099
Open this publication in new window or tab >>Proceedings of the 1st International Workshop on Cognition and Artificial Intelligence for Human-Centred Design 2017 co-located with IJCAI 2017
2018 (English)Conference proceedings (editor) (Other academic)
Place, publisher, year, edition, pages
Technical University of Aachen, 2018
Series
CEUR Workshop Proceedings, E-ISSN 1613-0073 ; 2099
National Category
Computer Sciences Design Human Computer Interaction Psychology
Identifiers
urn:nbn:se:oru:diva-69935 (URN)
Conference
1st International Workshop on Cognition and Artificial Intelligence for Human-Centred Design (CAID), Melbourne, Australia, August 19, 2017
Available from: 2018-10-29 Created: 2018-10-29 Last updated: 2018-11-01Bibliographically approved
Organisations
Identifiers
ORCID iD: ORCID iD iconorcid.org/0000-0002-6290-5492

Search in DiVA

Show all publications