The paper illustrates the utility to switch from a 3D content model to a Historic Building Information Modelling (HBIM) in order to support conservation and management of built heritage. This three dimensional solution is based on simplified parametric models, suitable for industrial elements and modern architecture, that can be usefully applied to heritage documentation and management of the data on conservation practices. In this sense, the potentials in starting the definition of an HBIM targeted library are investigated, towards the logic of object data definition, beginning from surface surveying and representation. In order to motivate the opportunity in using this 3D object modelling instruments, some case studies are investigated in the paper. Vault and wooden bean floor analysis show how a HBIM for architectural heritage could be implemented in order to assemble different kind of data on historical buildings, such as e.g. dimensional, geometrical, thematic, historical and architectural information.

Although aerial image-based bathymetric mapping can provide, unlike acoustic or LiDAR (Light Detection and Ranging) sensors, both water depth and visual information, water refraction poses significant challenges for accurate depth estimation. In order to tackle this challenge, we propose an image correction methodology, which first exploits recent machine learning procedures that recover depth from image-based dense point clouds and then corrects refraction on the original imaging dataset. This way, the structure from motion (SfM) and multi-view stereo (MVS) processing pipelines are executed on a refraction-free set of aerial datasets, resulting in highly accurate bathymetric maps. Performed experiments and validation were based on datasets acquired during optimal sea state conditions and derived from four different test-sites characterized by excellent sea bottom visibility and textured seabed. Results demonstrated the high potential of our approach, both in terms of bathymetric accuracy, as well as texture and orthoimage quality.

The determination of accurate bathymetric information is a key element for near offshore activities; hydrological studies, such as coastal engineering applications, sedimentary processes, hydrographic surveying, archaeological mapping and biological research. Through structure from motion (SfM) and multi-view-stereo (MVS) techniques, aerial imagery can provide a low-cost alternative compared to bathymetric LiDAR (Light Detection and Ranging) surveys, as it offers additional important visual information and higher spatial resolution. Nevertheless, water refraction poses significant challenges on depth determination. Till now, this problem has been addressed through customized image-based refraction correction algorithms or by modifying the collinearity equation. In this article, in order to overcome the water refraction errors in a massive and accurate way, we employ machine learning tools, which are able to learn the systematic underestimation of the estimated depths. In particular, an SVR (support vector regression) model was developed, based on known depth observations from bathymetric LiDAR surveys, which is able to accurately recover bathymetry from point clouds derived from SfM-MVS procedures. Experimental results and validation were based on datasets derived from different test-sites, and demonstrated the high potential of our approach. Moreover, we exploited the fusion of LiDAR and image-based point clouds towards addressing challenges of both modalities in problematic areas.

In this paper, an exhaustive review and comprehensive analysis of recent and former deep learning methods in 3D semantic segmentation (3DSS) is presented. In the related literature, the taxonomy scheme used for the classification of 3DSS deep learning methods is ambiguous. Based on the taxonomy schemes of nine existing review papers, a new taxonomy scheme for 3DSS deep learning methods is proposed, aiming to standardize it and improve the comparability and clarity across related studies. Furthermore, an extensive overview of the available 3DSS indoor and outdoor datasets is provided along with their links. The core part of this review is the detailed presentation of recent and former 3DSS deep learning methods and their classification using the proposed taxonomy scheme along with their GitHub repositories. Additionally, a brief but informative analysis of the evaluation metrics and loss functions used in 3DSS is included. Finally, a fruitful discussion of the examined 3DSS methods and datasets is presented to foster new research directions and applications in the field of 3DSS. In addition to this review, a GitHub repository is provided, including an initial classification of over 400 3DSS methods, using the proposed taxonomy scheme.

Conventional multi-view stereo (MVS) approaches based on photo-consistency measures are generally robust, yet often fail in calculating valid depth pixel estimates in low textured areas of the scene. In this study, a novel approach is proposed to tackle this challenge by leveraging semantic priors into a PatchMatch-based MVS in order to increase confidence and support depth and normal map estimation. Semantic class labels on image pixels are used to impose class-specific geometric constraints during multiview stereo, optimising the depth estimation on weakly supported, textureless areas, commonly present in urban scenarios of building facades, indoor scenes, or aerial datasets. Detecting dominant shapes, e.g., planes, with RANSAC, an adjusted cost function is introduced that combines and weighs both photometric and semantic scores propagating, thus, more accurate depth estimates. Being adaptive, it fills in apparent information gaps and smoothing local roughness in problematic regions while at the same time preserves important details. Experiments on benchmark and custom datasets demonstrate the effectiveness of the presented approach.

This paper focuses on the integration of multi-sensor techniques regarding the acquisition, processing, visualisation and management of data regarding historic stone structures. The interdisciplinary methodology that is carried out here comprises of two parts. In the first part, the acquisition of qualitative and quantitative data concerning the geometry, the materials and the degradation of the tangible heritage asset each time, is discussed. The second part, refers to the analysis, management and visualization of the interrelated data by using spatial information technologies. Through the paradigm of the surveying of the ancient temple of Pythian Apollo at Acropolis of Rhodes, Rhodes Island, Greece, it is aimed to highlight the issues deriving from the separate application of documentation procedures and how the fusion of these methods can contribute effectively to ensure the completeness of the measurements for complex structures. The surveying results are further processed to be compatible and integrated with GIS. Also, the geometric documentation derivatives are combined with environmental data and the results of the application of non-destructive testing and evaluation techniques in situ and analytical techniques in lab after sampling. GIS operations are utilized to document the building materials but also to model and to analyse the decay extent and patterns. Detailed surface measurements and geo-processing analysis are executed. This integrated approach, helps the assessment of past interventions on the monument, identify main causes of damage and decay, and finally assist the decision making on the most compatible materials and techniques for protection and restoration works.