Thanasis Zoumpekas

Selected Publications

1. 

   Machining tool identification utilizing temporal 3D point clouds

   Thanasis Zoumpekas, Alexander Leutgeb, Anna Puig, and Maria Salamó

   Journal of Intelligent Manufacturing 2023

   Abs Bib PDF Code

   The manufacturing domain is regarded as one of the most important engineering areas. Recently, smart manufacturing merges the use of sensors, intelligent controls, and software to manage each stage in the manufacturing lifecycle. Additionally, the increasing use of point clouds to model real products and machining tools in a virtual space facilitates the more accurate monitoring of the end-to-end production lifecycle. Thus, the conjunction of both, intelligent methods and more accurate 3D models allows the prediction of uncertainties and anomalies in the manufacturing process as well as reduces the final production costs. However, the high complexity of the geometrical structures defined by point clouds and the high accuracy required by the Quality Assurance/Quality control parameters during the process, pave the way for continuous improvements in smart manufacturing methods. This paper addresses a comprehensive analysis of machining tool identification utilizing temporal point cloud data. Specifically, we deal with the identification of machining tools from temporal 3D point clouds. To do that, we propose a process to construct and train intelligent models utilizing such data. Moreover, in our case study, we provide the research community with two labeled temporal 3D point cloud datasets, and we experiment with the pioneering PointNet neural network and three of its variants demonstrating an accuracy of 95% in the identification of the utilized machining tools in a machining process. Finally, we provide a prototype end-to-end intelligent system of machining tool identification.

   @article{Zoumpekas2023MachiningTools,
     title = {{Machining tool identification utilizing temporal 3D point clouds}},
     year = {2023},
     journal = {Journal of Intelligent Manufacturing},
     author = {Zoumpekas, Thanasis and Leutgeb, Alexander and Puig, Anna and Salam{\'{o}}, Maria},
     publisher = {Springer},
     url = {https://doi.org/10.1007/s10845-023-02093-5},
     doi = {10.1007/s10845-023-02093-5},
     issn = {1572-8145},
   }

2. 

   Rethinking Design and Evaluation of 3D Point Cloud Segmentation Models

   Thanasis Zoumpekas, Maria Salamó, and Anna Puig

   Remote Sensing 2022

   Abs Bib PDF

   Currently, the use of 3D point clouds is rapidly increasing in many engineering fields, such as geoscience and manufacturing. Various studies have developed intelligent segmentation models providing accurate results, while only a few of them provide additional insights into the efficiency and robustness of their proposed models. The process of segmentation in the image domain has been studied to a great extent and the research findings are tremendous. However, the segmentation analysis with point clouds is considered particularly challenging due to their unordered and irregular nature. Additionally, solving downstream tasks with 3D point clouds is computationally inefficient, as point clouds normally consist of thousands or millions of points sparsely distributed in 3D space. Thus, there is a significant need for rigorous evaluation of the design characteristics of segmentation models, to be effective and practical. Consequently, in this paper, an in-depth analysis of five fundamental and representative deep learning models for 3D point cloud segmentation is presented. Specifically, we investigate multiple experimental dimensions, such as accuracy, efficiency, and robustness in part segmentation (ShapeNet) and scene segmentation (S3DIS), to assess the effective utilization of the models. Moreover, we create a correspondence between their design properties and experimental properties. For example, we show that convolution-based models that incorporate adaptive weight or position pooling local aggregation operations achieve superior accuracy and robustness to point-wise MLPs, while the latter ones show higher efficiency in time and memory allocation. Our findings pave the way for an effective 3D point cloud segmentation model selection and enlighten the research on point clouds and deep learning.

   @article{Zoumpekas2022RethinkingModels,
     title = {{Rethinking Design and Evaluation of 3D Point Cloud Segmentation Models}},
     year = {2022},
     journal = {Remote Sensing},
     author = {Zoumpekas, Thanasis and Salam{\'{o}}, Maria and Puig, Anna},
     publisher = {Multidisciplinary Digital Publishing Institute},
     url = {https://doi.org/10.3390/RS14236049},
     doi = {10.3390/RS14236049},
     issn = {2072-4292},
   }

3. 

   Machine Learning-Based Rockfalls Detection with 3D Point Clouds, Example in the Montserrat Massif (Spain)

   Laura Blanco, David García-Sellés, Marta Guinau, Thanasis Zoumpekas, Anna Puig, Maria Salamó, Oscar Gratacós, Josep Anton Muñoz, Marc Janeras, and Oriol Pedraza

   Remote Sensing 2022

   Abs Bib PDF

   Rock slope monitoring using 3D point cloud data allows the creation of rockfall inventories, provided that an efficient methodology is available to quantify the activity. However, monitoring with high temporal and spatial resolution entails the processing of a great volume of data, which can become a problem for the processing system. The standard methodology for monitoring includes the steps of data capture, point cloud alignment, the measure of differences, clustering differences, and identification of rockfalls. In this article, we propose a new methodology adapted from existing algorithms (multiscale model to model cloud comparison and density-based spatial clustering of applications with noise algorithm) and machine learning techniques to facilitate the identification of rockfalls from compared temporary 3D point clouds, possibly the step with most user interpretation. Point clouds are processed to generate 33 new features related to the rock cliff differences, predominant differences, or orientation for classification with 11 machine learning models, combined with 2 undersampling and 13 oversampling methods. The proposed methodology is divided into two software packages: point cloud monitoring and cluster classification. The prediction model applied in two study cases in the Montserrat conglomeratic massif (Barcelona, Spain) reveal that a reduction of 98% in the initial number of clusters is sufficient to identify the totality of rockfalls in the first case study. The second case study requires a 96% reduction to identify 90% of the rockfalls, suggesting that the homogeneity of the rockfall characteristics is a key factor for the correct prediction of the machine learning models.

   @article{Blanco2022RemoteSensing,
     author = {Blanco, Laura and García-Sellés, David and Guinau, Marta and Zoumpekas, Thanasis and Puig, Anna and Salamó, Maria and Gratacós, Oscar and Muñoz, Josep Anton and Janeras, Marc and Pedraza, Oriol},
     title = {Machine Learning-Based Rockfalls Detection with 3D Point Clouds, Example in the Montserrat Massif (Spain)},
     journal = {Remote Sensing},
     volume = {14},
     year = {2022},
     number = {17},
     article-number = {4306},
     issn = {2072-4292},
     doi = {10.3390/rs14174306},
   }

4. Plotly 500 Index

   CLOSED: A Dashboard for 3D Point Cloud Segmentation Analysis using Deep Learning

   Thanasis Zoumpekas, Guillem Molina, Anna Puig, and Maria Salamó

   In Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP, 2022

   Abs Bib PDF Code Poster

   With the growing interest in 3D point cloud data, which is a set of data points in space used to describe a 3D object, and the inherent need to analyze it using deep neural networks, the visualization of data processes is critical for extracting meaningful insights. There is a gap in the literature for a full-suite visualization tool to analyse 3D deep learning segmentation models on point cloud data. This paper proposes such a tool to cover this gap, entitled point CLOud SEgmentation Dashboard (CLOSED). Specifically, we concentrate our efforts on 3D point cloud part segmentation, where the entire shape and the parts of a 3D object are significant. Our approach manages to (i) exhibit the learning evolution of neural networks, (ii) compare and evaluate different neural networks, (iii) highlight key-points of the segmentation process. We illustrate our proposal by analysing five neural networks utilizing the ShapeNet-part dataset.

   @inproceedings{Zoumpekas2022VISAPP,
     author = {Zoumpekas, Thanasis and Molina, Guillem and Puig, Anna and Salamó, Maria},
     title = {CLOSED: A Dashboard for 3D Point Cloud Segmentation Analysis using Deep Learning},
     booktitle = {Proceedings of the 17th International Joint Conference on Computer Vision, Imaging and Computer Graphics Theory and Applications - Volume 4: VISAPP,},
     year = {2022},
     pages = {403-410},
     publisher = {SciTePress},
     organization = {INSTICC},
     doi = {10.5220/0010826000003124},
     isbn = {978-989-758-555-5},
     issn = {2184-4321},
   }

5. Best Paper Award

   Benchmarking Deep Learning Models on Point Cloud Segmentation

   Thanasis Zoumpekas, Guillem Molina, Maria Salamó, and Anna Puig

   In Artificial Intelligence Research and Development 2021

   Abs Bib PDF

   Point clouds are currently used for a variety of applications, such as detection tasks in medical and geological domains. Intelligent analysis of point clouds is considered a highly computationally demanding and challenging task, especially the segmentation task among the points. Although numerous deep learning models have recently been proposed to segment point cloud data, there is no clear instruction of which exactly neural network to utilize and then incorporate into a system dealing with point cloud segmentation analysis. Besides, the majority of the developed models emphasize more on the accuracy rather than the efficiency, in order to achieve great results. Consequently, the training, validation and testing phases of the models require a great number of processing hours and a huge amount of memory. These high computational requirements are commonly difficult to deal with for many users. In this article, we analyse five state-of-the-art deep learning models for part segmentation task and give meaningful insights into the utilization of each one. We advance guidelines based on different properties, considering both learning-related metrics, such as accuracy, and system-related metrics, such as run time and memory footprint. We further propose and analyse generalized performance metrics, which facilitate the model evaluation phase in segmentation tasks allowing users to select the most appropriate approach for their context in terms of accuracy and efficiency.

   @incollection{Zoumpekas2021CCIA,
     title = {{Benchmarking Deep Learning Models on Point Cloud Segmentation}},
     year = {2021},
     booktitle = {Artificial Intelligence Research and Development},
     series = {Frontiers in Artificial Intelligence and Applications},
     author = {Zoumpekas, Thanasis and Molina, Guillem and Salam{\'{o}}, Maria and Puig, Anna},
     pages = {335--344},
     volume = {339},
     doi = {10.3233/FAIA210152},
   }

6. 

   An intelligent framework for end-to-end rockfall detection

   Thanasis Zoumpekas, Anna Puig, Maria Salamó, David Garcı́a-Sellés, Laura Blanco Nuñez, and Marta Guinau

   International Journal of Intelligent Systems 2021

   Abs Bib PDF

   Rockfall detection is a crucial procedure in the field of geology, which helps to reduce the associated risks. Currently, geologists identify rockfall events almost manually utilizing point cloud and imagery data obtained from different caption devices such as Terrestrial Laser Scanner (TLS) or digital cameras. Multitemporal comparison of the point clouds obtained with these techniques requires a tedious visual inspection to identify rockfall events which implies inaccuracies that depend on several factors such as human expertize and the sensibility of the sensors. This paper addresses this issue and provides an intelligent framework for rockfall event detection for any individual working in the intersection of the geology domain and decision support systems. The development of such an analysis framework presents major research challenges and justifies exhaustive experimental analysis. In particular, we propose an intelligent system that utilizes multiple machine learning algorithms to detect rockfall clusters of point cloud data. Due to the extremely imbalanced nature of the problem, a plethora of state-of-the-art resampling techniques accompanied by multiple models and feature selection procedures are being investigated. Various machine learning pipeline combinations have been examined and benchmarked applying well-known metrics to be incorporated into our system. Specifically, we developed machine learning techniques and applied them to analyze point cloud data extracted from TLS in two distinct case studies, involving different geological contexts: the basaltic cliff of Castellfollit de la Roca and the conglomerate Montserrat Massif, both located in Spain. Our experimental results indicate that some of the above-mentioned machine learning pipelines can be utilized to detect rockfall incidents on mountain walls, with experimentally validated accuracy.

   @article{Zoumpekas2021IntellSys,
     author = {Zoumpekas, Thanasis and Puig, Anna and Salamó, Maria and Garcı́a-Sellés, David and Blanco Nuñez, Laura and Guinau, Marta},
     title = {An intelligent framework for end-to-end rockfall detection},
     journal = {International Journal of Intelligent Systems},
     volume = {36},
     number = {11},
     pages = {6471-6502},
     doi = {https://doi.org/10.1002/int.22557},
     year = {2021},
   }