Yamini Patankar


Last Name

Patankar

First Name

Yamini

ORCID

0009-0007-4288-7809

Organisational unit

03891 - Flatt, Robert J. / Flatt, Robert J.

Filters

2024 - 20265
Reset filters

Search Results

Publications1 - 5 of 5
  • Sustainable Digital Concrete: Myth, Reality or Emerging Opportunity?
    Item type: Journal Article
    Wangler, Timothy; Patankar, Yamini; Flatt, Robert J. (2024)
    Architectural Design
    Unbelievably huge amounts of concrete, a major contributor to our contemporary world, are produced every year, and this continues to increase exponentially. ETH Zurich researchers Timothy Wangler, Yamini Patankar and Robert J Flatt explain the pros and cons of the digital fabrication of concrete, and the research still to be done in this relatively young and experimental subset of the construction industry and material science.
  • A 3D Model with Extended Reality of Lausanne Cathedral
    Item type: Conference Paper
    Varga, Matej; Patankar, Yamini; Grossenbacher, Nando; et al. (2024)
    SIG 2024 - Proceedings of the International Symposium on Engineering Geodesy
    The field of 3D modeling, encompassing both geometry and texture information, has been rapidly advancing, particularly with the integration of semantic information, extended reality (XR), and Building Information Modeling (BIM) capabilities. These content-rich digital models expand the scope of applications in various fields, including virtual tours, digital reconstruction, and heritage preservation. Traditionally, data collection has been a core task of geomatics, but efficient data processing, modeling, and visualization require interdisciplinary expertise. This paper presents a workflow detailing all steps from data planning and collection to the creation of a 3D model enriched with virtual capabilities, using the Lausanne Cathedral in Switzerland as a case study. Data collection was performed using terrestrial, handheld, and mobile laser scanners and RGB cameras. We discuss our decisions to ensure high-quality data collection, addressing variables such as overlap, challenging environments, and noise control. We detail the steps to reconstitute the cathedral's geometry, including registration, noise cleaning, subsampling, and addressing bottlenecks at each stage. Finally, we describe the development of VR experiences for various cathedral spaces, presenting selected case examples.
  • Heritage ++, a Spatial Computing approach to Heritage Conservation
    Item type: Journal Article
    Patankar, Yamini; Tennenini, Camilla; Bischof, Rafael; et al. (2024)
    RILEM Technical Letters
    Historic structures are affected by numerous degradation processes driven by a complex system of interconnected and mutually influencing factors. Preserving these monuments is a multidisciplinary endeavour that extends beyond one-time interventions, necessitating a comprehensive methodology that involves various stakeholders, expert consultations, monitoring tools, and impact assessments. Limitations arise due to communication barriers and difficulty in translating and transferring experience among disciplines, often compromising the collective ability to define the best possible conservation strategies.Recent advancements in 3D modelling and data management technologies offer collaborative platforms for information sharing. However, the complex interfaces of these tools often limit their accessibility, making them exclusive to specialists. Integrating Spatial Computing could address these challenges by fostering intuitive engagement and enhancing accessibility and depth in interdisciplinary interactions. This letter outlines initial efforts in using spatial computing to tackle the challenges of built heritage conservation and presents a vision for its future development.
  • Immersive Conservation: Interactive Tools for Preserving Built Heritage
    Item type: Conference Paper
    Patankar, Yamini; Mitterberger, Daniela; Brahimsamba, Bomou; et al. (2026)
    STONE 2025 : Proceedings of the 15th International Congress on the Deterioration and Conservation of Stone, Volume 1 & 2
    The conservation of stone-built heritage presents unique challenges, ranging from documenting complex geometries and understanding the underlying damage mechanisms to implementing conservation strategies. Addressing these challenges requires a comprehensive methodology involving a multidisciplinary team of stakeholders and experts, monitoring tools, and impact assessments. Recent advancements in Heritage Building Information Modelling (HBIM) offer novel solutions for the integration of datasets from multiple disciplines to enhance collaboration among experts. However, HBIM models are often complex, static, and not connected to real-time data, limiting their applicability to a single case study and preventing broader use across multiple contexts. Extended Reality (XR) technologies offer much-needed solutions for enhancing visualisation and intuitive engagement with these models. This paper introduces a novel framework that combines HBIM and XR to address stone conservation challenges. HBIM serves as a robust platform for 3D modelling and data integration, while XR enables spatial visualisation and interactive engagement, facilitating both on-site and remote expert consultations. The preliminary framework is developed for a reference monument, the Lausanne Cathedral. This integrated approach can foster collaboration across disciplines, allowing stakeholders to access the information in the HBIM model, and overcoming the compartmentalisation of data.
  • Machine Learning-Based Detection of Stone Degradation using TLS, Photographs and HBIM: A Case Study on the Lausanne Cathedral
    Item type: Conference Paper
    Laasch, Helena; Patankar, Yamini; Tennenini, Camilla; et al. (2026)
    STONE 2025: Proceedings of the 15th International Congress on the Deterioration and Conservation of Stone, Volume 1 & 2
    Historic sandstone structures are vulnerable to environmental degradation, requiring efficient damage assessment and documentation for effective conservation. Traditional manual mapping methods are time-intensive, cost-intensive, and challenging to standardize. Herein we present a semi-automated workflow for damage classification and integration into a historic building information model (HBIM) using terrestrial laser scanning (TLS), orthophotos, and machine learning. A Random Forest model, optimised through forward feature selection and Bayesian hyperparameter tuning, classifies degradation types (contour scaling, biodeterioration, black crust formation and exfoliation) based on geometric and radiometric features. Finally, degradation information is linked to individual stone blocks in the HBIM, and degradation maps can be generated. In a case study on the Lausanne Cathedral, we achieve an overall classification accuracy of 80% using this approach. The results highlight the potential of using machine-learning techniques with TLS data for an efficient and scalable heritage condition assessment.
Publications1 - 5 of 5