Bryan German Pantoja Rosero

Verified

Bryan verified their affiliation via an institutional email.

Learn more

Verified

Bryan verified their affiliation via an institutional email.

Learn more

• PhD
• PostDoc Position at University of Texas at Austin

This paper describes a pipeline for automatically generating level of detail (LOD) models (digital twins), specifically LOD2 and LOD3, from free-standing buildings. Our approach combines structure from motion (SfM) with deep-learning-based segmentation techniques. Given multiple-view images of a building, we compute a three-dimensional (3D) planar...

Current procedures for the rapid inspection of buildings and infrastructure are subjective, time-consuming, and cumbersome to document, necessitating new technologies to automate the process and eliminate these shortcomings. Fortunately, recent developments in imaging devices and artificial intelligence, such as computer vision, provide the necessa...

To predict the response of masonry buildings to various types of loads, engineers use finite element models, specifically solid-element and macro-element models. For predicting masonry responses to seismic events in particular, equivalent frame models—a subcategory of macro-element models—are a common choice because of their low computational cost....

Our research introduces a novel framework for generating detailed 3D building models (LOD4) that integrates both exterior and interior data. This approach addresses a significant gap in current methods that focus primarily on either the interior or exterior of buildings. By leveraging structure-from-motion (SfM) models, planar primitives, and image...

We introduce a new approach to active learning that addresses the inefficiencies of traditional training methods. Rather than requiring full annotations for all images, our method leverages partial annotations guided by predictions and uncertainty measures from previously trained models. This approach reduces labeling effort and training time witho...

Recent advancements in deep learning have found compelling applications in the image-based inspection of infrastructure systems. However, the efficacy of these models hinges significantly on the quality and diversity of the input data used for training. In this study, we rigorously evaluate the performance of a convolutional neural network architec...

Neural radiance fields have emerged as a dominant paradigm for creating complex 3D environments incorporating synthetic novel views. However, 3D object removal applications utilizing neural radiance fields have lagged behind in effectiveness, particularly when open set queries are necessary for determining the relevant objects. One such application...

The use of natural stones as building material can help reducing the carbon footprint of the construction industry. However, their non-uniform shapes makes the construction of stone masonry structures challenging. Therefore, the development of efficient algorithms for the stacking of irregular stones obeying structural and architectonic requirement...

Image information about the state of a building after an earthquake, which can be collected without endangering the post-earthquake reconnaissance activities, can be used to reduce uncertainties in response predictions for future seismic events. This paper investigates the impact of using data from image-based inspection of building facades on redu...

Digital twins are virtual models of physical objects or systems that enable real-time monitoring and analysis. In the field of stone masonry buildings, digital twins can be used to assess damage, predict maintenance needs, and optimize building performance. However, creating and analyzing digital twins of stone masonry buildings can be a complex an...

We present an image-based pipeline for generating geometrical digital twins (GDTs) of stone masonry elements with detail down to the stone level. For this purpose, we acquire RGB images of the individual stones and of the wall during the construction phase. In our framework, we use structure from motion (SfM) to first generate 3D source and destina...

We present a method for segmenting cracks in images of masonry buildings damaged by earthquakes. Existing methods of crack detection fail to preserve the continuity of cracks, and their performance deteriorates with imprecise training labels. We address these problems by adapting an approach previously proposed for reconstructing roads in aerial im...

Determining the relationship between the cause of damage and the subsequent structural behavior of infrastructure systems requires an accurate characterization of the propagation of cracks, which represents the evolution of the damage state. When no information about the cause of damage is available, kinematic approaches can be used to describe the...

The structures of reinforced concrete when submitted to mechanical stresses present non-linear behaviour due to different chemical and / or physical phenomena. Among these phenomena are, for example, the cracking of the concrete given the tensile stresses, crushing due to the stresses of compression, creep, etc. This causes the concrete to have dam...

The concrete, because it is a fragile material, presents behaviours related to permanent deformations and losses of the rigidity of the structures. These behaviours concern different phenomena such as, for example, the damage that the material suffers. On the other hand, there are different constitutive models that simulate the tensional and deform...

Currently, studies of the mechanical behavior under dynamic loading of reinforced concrete structures through the Finite Element Method (FEM) present difficulties in the analysis of the reinforcement. This is due to the limitations of the methodologies used in the modeling of the rebars in the FEM, such as the location of the elements that represen...