Space–use analysis through computer vision

Tomé, Ana; Kuipers, Martijn; Pinheiro, Tiago; Nunes, Mário Serafim; Heitor, Teresa

doi:10.1016/j.autcon.2015.04.013

Cited by 14 publications

(9 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Metrics such as the percentage of desks occupied in a workspace can be used to determine the overall spatial efficiency [114]. With new methods enabling real-time, detailed inference of occupants' space utilization [115][116][117], researchers have defined metrics that explore the potential to improve overall space utilization rates by moving to a scenario in which occupants share desks [35].…”

Section: Space Planning and Organizational Metricsmentioning

confidence: 99%

Simulation-aided occupant-centric building design: A critical review of tools, methods, and applications

Azar

O’Brien

Carlucci

et al. 2020

Energy and Buildings

View full text Add to dashboard Cite

Occupants are active participants in their built environment, affecting its performance while simultaneously being affected by its design and indoor environmental conditions. With recent advances in computer modeling, simulation tools, and analysis techniques, topics such as human-building interactions and occupant behavior have gained significant interest in the literature given their premise of improving building design processes and operating strategies. In practice, the focus of occupant-centric literature has been mostly geared towards the latter (i.e., operation), leaving the implications on building design practices underexplored. This paper fills the gap by providing a critical review of existing studies applying computerbased modeling and simulation to guide occupant-centric building design. The reviewed papers are organized along four main themes, namely occupant-centric: (i) metrics of building performance, (ii) modeling and simulation approaches, (iii) design methods and applications, and (iv) supporting practices and mechanisms. Important barriers are identified for a more effective application of occupant-centric building design practices including the limited consideration of metrics beyond energy efficiency (e.g., occupant well-being and space planning), the limited implementation and validation of the proposed methods, and the lack of integration of occupant behavior modeling in existing building performance simulation tools. Future research directions include the need for large-scale international data collection efforts to move from generic assumptions about occupant behavior to specific/localized knowledge, the need for improved metrics of measuring building performance, as well as the need for industry practices, such as building codes, to promote an occupant-in-the-loop approach to the building design process.

show abstract

Section: Space Planning and Organizational Metricsmentioning

confidence: 99%

Simulation-aided occupant-centric building design: A critical review of tools, methods, and applications

Azar

O’Brien

Carlucci

et al. 2020

Energy and Buildings

View full text Add to dashboard Cite

show abstract

“…Similarly, observational techniques such as video analysis (e.g. Tomé et al 2015), movement tracking (e.g. Tröndle et al 2014;Dalton et al 2012) and virtual reality simulations (Conroy Dalton 2001) have been employed, increasingly relying on usability metrics (e.g.…”

Section: Resultsmentioning

confidence: 99%

Applying HCI Methods and Concepts to Architectural Design (Or Why Architects Could Use HCI Even If They Don’t Know It)

Krukar

Dalton

Hölscher

2016

Human–Computer Interaction Series

View full text Add to dashboard Cite

The act of designing a building is indirectly, but conceptually very closely, linked to the user experience of its final outcome. It is this experience which often constitutes a major criterion for assessing the quality of the architect's work. And yet, it would be a gross overstatement to suggest that architectural design is a user-centered process.On a more generic level, designing any physical object acting as a catalyst for the final experience can be viewed as an act of designing a human-artifact interaction where the 'artifact' (be it a building or a computer device) serves as an interface for the ultimate behavior or emotional reaction. This chapter argues, that the field of Human-Computer Interaction (HCI) can be viewed as a source of inspiration for architects wishing to incorporate, or enhance, user-centric planning routines in their creative workflows.Drawing from the methodological toolbox of HCI, we demonstrate how usercentric planning can be placed in a structured framework, with tested and easy-toapply methods serving as the vehicle for holistic user-centered planning processes.The chapter proposes a formal model for understanding usability and user experience in the architectural context, demonstrates a number of methods suitable for its application, and concludes with a case study of an attempted use of one of such methods in an award-winning (yet, not necessarily user-friendly) public library project.

show abstract

“…In the future, we will investigate to release this QP problem to a linear problem, by using the expectation-maximization (EM) framework to release the hinge loss to a linear function. Moreover, we also plan to extend the proposed algorithm to different applications, e.g., bioinformatics [43,34,36,47], computer vision [32,37,6,31], and information retrieval [42,16,11,33].…”

Section: Discussionmentioning

confidence: 99%

Supervised learning of sparse context reconstruction coefficients for data representation and classification

Liu

Wang

Yin

et al. 2015

Neural Comput & Applic

View full text Add to dashboard Cite

Context of data points, which is usually defined as the other data points in a data set, has been found to paly important roles in data representation and classification. In this paper, we study the problem of using context of a data point for its classification problem. Our work is inspired by the observation that actually only very few data points are critical in the context of a data point for its representation and classification. We propose to represent a data point as the sparse linear combination of its context, and learn the sparse context in a supervised way to increase its discriminative ability. To this end, we proposed a novel formulation for context learning, by modeling the learning of context parameter and classifier in a unified objective, and optimizing it with an alternative strategy in an iterative algorithm. Experiments on three benchmark data set show its advantage over state-of-the-art context-based data representation and classification methods.

show abstract

Space–use analysis through computer vision

Cited by 14 publications

References 31 publications

Simulation-aided occupant-centric building design: A critical review of tools, methods, and applications

Simulation-aided occupant-centric building design: A critical review of tools, methods, and applications

Applying HCI Methods and Concepts to Architectural Design (Or Why Architects Could Use HCI Even If They Don’t Know It)

Supervised learning of sparse context reconstruction coefficients for data representation and classification

Contact Info

Product

Resources

About