Alberto Manuel Esmorís Pena
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Research Subjects
Alberto Manuel Esmorís Pena has a total of 17 co-authors. Here are the top co-authors by number of shared publications:
- Bernhard Höfle
DGeo Research Group, Institute of Geography Heidelberg University Heidelberg Germany
4 shared publications - Lukas Winiwarter
Unit of Geometry and Surveying, Faculty of Engineering Sciences, University of Innsbruck, Innsbruck, Austria
4 shared publications - Francisco Fernandez Rivera
Centro Singular de Investigación en Tecnoloxías Intelixentes (CiTIUS), University of Santiago de Compostela, 15782 Santiago de Compostela, Spain
3 shared publications Hannah Weiser
Institute of Geography 3DGeo Research Group Heidelberg University Im Neuenheimer Feld 368 69120 Heidelberg Baden-Württemberg Germany
2 shared publications- Hannah Weiser
DGeo Research Group, Institute of Geography, Heidelberg University, Germany
2 shared publications - Jose Carlos Cabaleiro
Centro Singular de Investigación en Tecnoloxías Intelixentes (CiTIUS), Dpto. Electrónica e Computación, Universidade de Santiago de Compostela, Santiago de Compostela, Spain
2 shared publications
Latest Publications
- Open Access
- 17 March 2026
- Published by Elsevier in Journal of Computational and Applied Mathematics
- Vol. 485
Introducing the Taubin-Weingarten algorithm to compute second-order geometric descriptors in 3D point clouds
- 17 March 2026
- Published by Elsevier in Journal of Computational and Applied Mathematics
- Vol. 485
Open AccessIntroducing the Taubin-Weingarten algorithm to compute second-order geometric descriptors in 3D point clouds
- Open Access
- 22 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 485-501
Deep Learning for Ultra-Large-Scale Semantic Segmentation of Geographic 3D Point Clouds With Missing Labels
Semantic segmentation of 3D point clouds is a critical task essential for research and industry in a wide variety of domains. Mostworks until nowuse datasets where the concept of large-scale varies between 1.17 and 9, 261 million points and 0.31 and 250 square kilometers. In this research, we push the limits of 3D deep learning by considering approximately 36, 369 million points and 29, 557 square kilometers in our case study. To the best of our knowledge, it is the largest geographic region ever classified with neural networks for 3D point clouds in scientific works. The main contributions of our research are: 1) The first published results on semantic segmentation of ultra-large-scale and low-resolution airborne laser scanning point clouds, 2) adapted neural networks to work under missing labels and high measurement error conditions, 3) the introduction of class ambiguity as a more robust uncertainty measurement compared to entropy, and 4) an open-source deep learning framework for 3D point clouds that enables the reproducibility of our experiments and further studies.
- 22 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 485-501
Open AccessDeep Learning for Ultra-Large-Scale Semantic Segmentation of Geographic 3D Point Clouds With Missing Labels
Semantic segmentation of 3D point clouds is a critical task essential for research and industry in a wide variety of domains. Mostworks until nowuse datasets where the concept of large-scale varies between 1.17 and 9, 261 million points and 0.31 and 250 square kilometers. In this research, we push the limits of 3D deep learning by considering approximately 36, 369 million points and 29, 557 square kilometers in our case study. To the best of our knowledge, it is the largest geographic region ever classified with neural networks for 3D point clouds in scientific works. The main contributions of our research are: 1) The first published results on semantic segmentation of ultra-large-scale and low-resolution airborne laser scanning point clouds, 2) adapted neural networks to work under missing labels and high measurement error conditions, 3) the introduction of class ambiguity as a more robust uncertainty measurement compared to entropy, and 4) an open-source deep learning framework for 3D point clouds that enables the reproducibility of our experiments and further studies.
- Open Access
- 13 July 2024
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 215, 192-213
Deep learning with simulated laser scanning data for 3D point cloud classification
- 13 July 2024
- Published by Elsevier in ISPRS Journal of Photogrammetry and Remote Sensing
- Vol. 215, 192-213
Open AccessDeep learning with simulated laser scanning data for 3D point cloud classification
- Open Access
- 14 June 2024
- Published by MDPI AG in Smart Cities
- Vol. 7 (3) , 1441-1461
Methodology for Identifying Optimal Pedestrian Paths in an Urban Environment: A Case Study of a School Environment in A Coruña, Spain
Improving urban mobility, especially pedestrian mobility, is a current challenge in virtually every city worldwide. To calculate the least-cost paths and safer, more efficient routes, it is necessary to understand the geometry of streets and their various elements accurately. In this study, we propose a semi-automatic methodology to assess the capacity of urban spaces to enable adequate pedestrian mobility. We employ various data sources, but primarily point clouds obtained through a mobile laser scanner (MLS), which provide a wealth of highly detailed information about the geometry of street elements. Our method allows us to characterize preferred pedestrian-traffic zones by segmenting crosswalks, delineating sidewalks, and identifying obstacles and impediments to walking in urban routes. Subsequently, we generate different displacement cost surfaces and identify the least-cost origin–destination paths. All these factors enable a detailed pedestrian mobility analysis, yielding results on a raster with a ground sampling distance (GSD) of 10 cm/pix. The method is validated through its application in a case study analyzing pedestrian mobility around an educational center in a purely urban area of A Coruña (Galicia, Spain). The segmentation model successfully identified all pedestrian crossings in the study area without false positives. Additionally, obstacle segmentation effectively identified urban elements and parked vehicles, providing crucial information to generate precise friction surfaces reflecting real environmental conditions. Furthermore, the generation of cumulative displacement cost surfaces allowed for identifying optimal routes for pedestrian movement, considering the presence of obstacles and the availability of traversable spaces. These surfaces provided a detailed representation of pedestrian mobility, highlighting significant variations in travel times, especially in areas with high obstacle density, where differences of up to 15% were observed. These results underscore the importance of considering obstacles’ existence and location when planning pedestrian routes, which can significantly influence travel times and route selection. We consider the capability to generate accurate cumulative cost surfaces to be a significant advantage, as it enables urban planners and local authorities to make informed decisions regarding the improvement of pedestrian infrastructure.
- 14 June 2024
- Published by MDPI AG in Smart Cities
- Vol. 7 (3) , 1441-1461
Open AccessMethodology for Identifying Optimal Pedestrian Paths in an Urban Environment: A Case Study of a School Environment in A Coruña, Spain
Improving urban mobility, especially pedestrian mobility, is a current challenge in virtually every city worldwide. To calculate the least-cost paths and safer, more efficient routes, it is necessary to understand the geometry of streets and their various elements accurately. In this study, we propose a semi-automatic methodology to assess the capacity of urban spaces to enable adequate pedestrian mobility. We employ various data sources, but primarily point clouds obtained through a mobile laser scanner (MLS), which provide a wealth of highly detailed information about the geometry of street elements. Our method allows us to characterize preferred pedestrian-traffic zones by segmenting crosswalks, delineating sidewalks, and identifying obstacles and impediments to walking in urban routes. Subsequently, we generate different displacement cost surfaces and identify the least-cost origin–destination paths. All these factors enable a detailed pedestrian mobility analysis, yielding results on a raster with a ground sampling distance (GSD) of 10 cm/pix. The method is validated through its application in a case study analyzing pedestrian mobility around an educational center in a purely urban area of A Coruña (Galicia, Spain). The segmentation model successfully identified all pedestrian crossings in the study area without false positives. Additionally, obstacle segmentation effectively identified urban elements and parked vehicles, providing crucial information to generate precise friction surfaces reflecting real environmental conditions. Furthermore, the generation of cumulative displacement cost surfaces allowed for identifying optimal routes for pedestrian movement, considering the presence of obstacles and the availability of traversable spaces. These surfaces provided a detailed representation of pedestrian mobility, highlighting significant variations in travel times, especially in areas with high obstacle density, where differences of up to 15% were observed. These results underscore the importance of considering obstacles’ existence and location when planning pedestrian routes, which can significantly influence travel times and route selection. We consider the capability to generate accurate cumulative cost surfaces to be a significant advantage, as it enables urban planners and local authorities to make informed decisions regarding the improvement of pedestrian infrastructure.
- Open Access
- 18 January 2024
- Published by California Digital Library (CDL) in EarthArXiv
Deep learning with simulated laser scanning data for 3D point cloud classification
Laser scanning is an active remote sensing technique applied in many disciplines to acquire state-of-the-art spatial measurements. Semantic labeling is often necessary to extract information from the raw point cloud. Deep learning methods constitute a data-hungry solution for the semantic segmentation of point clouds. In this work, we investigate the use of simulated laser scanning for training deep learning models, which are applied to real data subsequently. We show that training a deep learning model purely on virtual laser scanning data can produce results comparable to models trained on real data when evaluated on real data. For leaf-wood segmentation of trees, using the KPConv model trained with virtual data achieves 93.7% overall accuracy, while the model trained on real data reaches 94.7% overall accuracy. In urban contexts, a KPConv model trained on virtual data achieves 74.1% overall accuracy on real validation data, while the model trained on real data achieves 82.4%. Our models outperform the state-of-the-art model FSCT in terms of generalization to unseen real data as well as a baseline model trained on points randomly sampled from the tree mesh surface. From our results, we conclude that the combination of laser scanning simulation and deep learning is a cost-effective alternative to real data acquisition and manual labeling in the domain of geospatial point cloud analysis. The strengths of this approach are that a) a large amount of diverse laser scanning training data can be generated quickly and without the need for expensive equipment, b) the simulation configurations can be adapted so that the virtual training data have similar characteristics to the targeted real data, and c) the whole workflow can be automated through procedural scene generation.
- 18 January 2024
- Published by California Digital Library (CDL) in EarthArXiv
Open AccessDeep learning with simulated laser scanning data for 3D point cloud classification
Laser scanning is an active remote sensing technique applied in many disciplines to acquire state-of-the-art spatial measurements. Semantic labeling is often necessary to extract information from the raw point cloud. Deep learning methods constitute a data-hungry solution for the semantic segmentation of point clouds. In this work, we investigate the use of simulated laser scanning for training deep learning models, which are applied to real data subsequently. We show that training a deep learning model purely on virtual laser scanning data can produce results comparable to models trained on real data when evaluated on real data. For leaf-wood segmentation of trees, using the KPConv model trained with virtual data achieves 93.7% overall accuracy, while the model trained on real data reaches 94.7% overall accuracy. In urban contexts, a KPConv model trained on virtual data achieves 74.1% overall accuracy on real validation data, while the model trained on real data achieves 82.4%. Our models outperform the state-of-the-art model FSCT in terms of generalization to unseen real data as well as a baseline model trained on points randomly sampled from the tree mesh surface. From our results, we conclude that the combination of laser scanning simulation and deep learning is a cost-effective alternative to real data acquisition and manual labeling in the domain of geospatial point cloud analysis. The strengths of this approach are that a) a large amount of diverse laser scanning training data can be generated quickly and without the need for expensive equipment, b) the simulation configurations can be adapted so that the virtual training data have similar characteristics to the targeted real data, and c) the whole workflow can be automated through procedural scene generation.
- Open Access
- 30 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 105052-105073
Virtual LiDAR Simulation as a High Performance Computing Challenge: Toward HPC HELIOS++
The software HELIOS++ simulates the laser scanning of a given virtual scene that can be composed of different spatial primitives and 3D meshes with distinct granularity. The high computational cost of this type of simulation software demands efficient computational solutions. Classical solutions based on GPU are not well suited when irregular geometries compose the scene combining different primitives and physics models because they lead to different computation branches. In this paper, we explore the usage of parallelization strategies based on static and dynamic workload balancing and heuristic optimization strategies to speed up the ray tracing process based on a k-dimensional tree (KDT). Using HELIOS++ as our case study, we analyze the performance of our algorithms on different parallel computers, including the CESGA FinisTerrae-II supercomputer. There is a significant performance boost in all cases, with the decrease in computation time ranging from 89.5% to 99.4%. Our results show that the proposed algorithms can boost the performance of any software that relies heavily on a KDT or a similar data structure, as well as those that spend most of the time computing with only a few synchronization barriers. Hence, the algorithms presented in this paper improve performance, whether computed on personal computers or supercomputers.
- 30 September 2022
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 10, 105052-105073
Open AccessVirtual LiDAR Simulation as a High Performance Computing Challenge: Toward HPC HELIOS++
The software HELIOS++ simulates the laser scanning of a given virtual scene that can be composed of different spatial primitives and 3D meshes with distinct granularity. The high computational cost of this type of simulation software demands efficient computational solutions. Classical solutions based on GPU are not well suited when irregular geometries compose the scene combining different primitives and physics models because they lead to different computation branches. In this paper, we explore the usage of parallelization strategies based on static and dynamic workload balancing and heuristic optimization strategies to speed up the ray tracing process based on a k-dimensional tree (KDT). Using HELIOS++ as our case study, we analyze the performance of our algorithms on different parallel computers, including the CESGA FinisTerrae-II supercomputer. There is a significant performance boost in all cases, with the decrease in computation time ranging from 89.5% to 99.4%. Our results show that the proposed algorithms can boost the performance of any software that relies heavily on a KDT or a similar data structure, as well as those that spend most of the time computing with only a few synchronization barriers. Hence, the algorithms presented in this paper improve performance, whether computed on personal computers or supercomputers.
- Top Cited PapersOpen Access
- 1 February 2022
- Published by Elsevier in Remote Sensing of Environment
- Vol. 269
Virtual laser scanning with HELIOS++: A novel take on ray tracing-based simulation of topographic full-waveform 3D laser scanning
- 1 February 2022
- Published by Elsevier in Remote Sensing of Environment
- Vol. 269
Top Cited PapersOpen AccessVirtual laser scanning with HELIOS++: A novel take on ray tracing-based simulation of topographic full-waveform 3D laser scanning