Jean-José Orteu
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
Jean-José Orteu has a total of 147 co-authors. Here are the top co-authors by number of shared publications:
- Igor Jovančević
University of Montenegro, Computer Science Center, Faculty of Natural Sciences and Mathematics, Podgorica, Montenegro
15 shared publications - Laurent Robert
Université de Toulouse; INSA, UPS, Mines Albi, ISAE ; ICA (Institut Clément Ader); Campus Jarlard, F-81013 Albi, France
12 shared publications - Hubert W. Schreier
Correlated Solutions, Inc
10 shared publications - Michael A. Michael A.
University of South Carolina
10 shared publications Thierry Sentenac
CNRS, Mines Albi, UPS, INSA, ISAE-SUPAERO, Campus Jarlard, Institut Clément Ader, Université de Toulouse, Toulouse, France
8 shared publicationsLudovic Brèthes
KEONYS
4 shared publications
Latest Publications
- 23 March 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 35 (03) , 1-22
Industrial monocular depth estimation, semantic segmentation, and 6D pose estimation using foundational depth models in a data-scarce setting
Many industrial inspection tasks often require 3D information in order to be solved properly. Some examples include object pose estimation with 6 degrees of freedom (6D), conformity checks and volumetric measurements. However, obtaining accurate 3D or depth information about a scene in an industrial environment is in itself a considerably challenging task, especially in production of mechanical parts, where many reflective metallic surfaces are encountered. Reflections make it difficult or even impossible to employ specialized depth sensors such as LiDARs and stereo 3D cameras, or traditional depth estimation and 3D reconstruction algorithms. Conversely, the fact that industrial use cases often involve working with a very low volume of available data usually prohibits the training of custom deep learning models for depth estimation. As a solution, we demonstrate the viability of utilizing large, pre-trained foundational models for monocular depth estimation (such as Depth Anything variants) in industrial production. More specifically, we showcase successful applications of these models in the domain of 5-axis machining setup inspection. The proposed methodology enables adjusting the outputs of foundational depth models in a use case-specific setting, based on known values of 5-axis machine coordinates. Importantly, such adjustments are performed in a way that avoids fine-tuning of neural network parameters, and therefore our methodology can be applied even under extreme data scarcity conditions, using less than 50 images. Moreover, we showcase two downstream applications of the proposed monocular depth estimation approach - depth-based semantic segmentation of objects of interest, as well as depth-based 6D pose estimation. It is shown that the absolute relative error (AbsRel) of our depth estimations can be as low as 2.54% on average, which in turn leads to very precise semantic segmentation (89.4% IoU) and object pose estimations (6.39 mm ADD for objects with dimensions 200 mm×200 mm×60 mm). Lastly, the proposed segmentation and pose estimation methods exhibit excellent generalization capabilities, maintaining strong accuracy even when tested on images of completely new setups.
- 23 March 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 35 (03) , 1-22
Industrial monocular depth estimation, semantic segmentation, and 6D pose estimation using foundational depth models in a data-scarce setting
Many industrial inspection tasks often require 3D information in order to be solved properly. Some examples include object pose estimation with 6 degrees of freedom (6D), conformity checks and volumetric measurements. However, obtaining accurate 3D or depth information about a scene in an industrial environment is in itself a considerably challenging task, especially in production of mechanical parts, where many reflective metallic surfaces are encountered. Reflections make it difficult or even impossible to employ specialized depth sensors such as LiDARs and stereo 3D cameras, or traditional depth estimation and 3D reconstruction algorithms. Conversely, the fact that industrial use cases often involve working with a very low volume of available data usually prohibits the training of custom deep learning models for depth estimation. As a solution, we demonstrate the viability of utilizing large, pre-trained foundational models for monocular depth estimation (such as Depth Anything variants) in industrial production. More specifically, we showcase successful applications of these models in the domain of 5-axis machining setup inspection. The proposed methodology enables adjusting the outputs of foundational depth models in a use case-specific setting, based on known values of 5-axis machine coordinates. Importantly, such adjustments are performed in a way that avoids fine-tuning of neural network parameters, and therefore our methodology can be applied even under extreme data scarcity conditions, using less than 50 images. Moreover, we showcase two downstream applications of the proposed monocular depth estimation approach - depth-based semantic segmentation of objects of interest, as well as depth-based 6D pose estimation. It is shown that the absolute relative error (AbsRel) of our depth estimations can be as low as 2.54% on average, which in turn leads to very precise semantic segmentation (89.4% IoU) and object pose estimations (6.39 mm ADD for objects with dimensions 200 mm×200 mm×60 mm). Lastly, the proposed segmentation and pose estimation methods exhibit excellent generalization capabilities, maintaining strong accuracy even when tested on images of completely new setups.
- 17 February 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 35 (03) , 1-15
Data augmentation for fuselage panel inspection via 3D point cloud segmentation
Automated inspection of mechanical assemblies can be significantly improved through the application of deep learning techniques to three-dimensional (3D) point cloud segmentation. In industrial environments, however, the use of such models is often constrained by the limited availability of annotated datasets, caused by the high cost, time demands, and confidentiality of data acquisition. To address these limitations, a fully instance and semantically annotated dataset of airplane fuselage panels collected in laboratory conditions is presented, together with a deep learning approach for detecting and classifying rivets in 3D scans. The proposed method enables the detection of defects such as missing or damaged rivets. To overcome limited training data, an augmentation strategy was deployed, designed to improve the detection of damaged and missing rivets. The influence of different scanning devices and resolutions on inspection accuracy is also evaluated across multiple deep learning models. This whole process was intended to be used as a preparation step in the reverse engineering pipeline.
- 17 February 2026
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 35 (03) , 1-15
Data augmentation for fuselage panel inspection via 3D point cloud segmentation
Automated inspection of mechanical assemblies can be significantly improved through the application of deep learning techniques to three-dimensional (3D) point cloud segmentation. In industrial environments, however, the use of such models is often constrained by the limited availability of annotated datasets, caused by the high cost, time demands, and confidentiality of data acquisition. To address these limitations, a fully instance and semantically annotated dataset of airplane fuselage panels collected in laboratory conditions is presented, together with a deep learning approach for detecting and classifying rivets in 3D scans. The proposed method enables the detection of defects such as missing or damaged rivets. To overcome limited training data, an augmentation strategy was deployed, designed to improve the detection of damaged and missing rivets. The influence of different scanning devices and resolutions on inspection accuracy is also evaluated across multiple deep learning models. This whole process was intended to be used as a preparation step in the reverse engineering pipeline.
- 15 August 2025
- Published by SPIE-Intl Soc Optical Eng
- Vol. 13737, 21-89
Detection and characterization of damages on mechanical parts via 3D point cloud processing
This paper presents an approach for defect detection and characterization on aeronautical metallic or composite surfaces, by processing 3D point clouds. For defect detection, we employed deep learning techniques, specifically segmentation models. To obtain a sufficient amount of training data, we developed a custom pipeline for generating synthetic point clouds that represent surfaces with various types of damage. In the testing phase, we used only real point clouds collected using a 3D scanner. For the characterization step, our system aims to provide precise measurements of the geometric properties of the detected defects, such as depth and surface area. For measuring the width and length of the defects, we utilized segmentation masks generated during the detection phase. To address the challenge of measuring defect depth, we relied on traditional computer vision techniques applied to 3D point clouds. We reconstructed an ideal, defect-free surface from points classified as non-defective, and then determined the defect depth as the maximum distance between the defective points and this reconstructed surface.
- 15 August 2025
- Published by SPIE-Intl Soc Optical Eng
- Vol. 13737, 21-89
Detection and characterization of damages on mechanical parts via 3D point cloud processing
This paper presents an approach for defect detection and characterization on aeronautical metallic or composite surfaces, by processing 3D point clouds. For defect detection, we employed deep learning techniques, specifically segmentation models. To obtain a sufficient amount of training data, we developed a custom pipeline for generating synthetic point clouds that represent surfaces with various types of damage. In the testing phase, we used only real point clouds collected using a 3D scanner. For the characterization step, our system aims to provide precise measurements of the geometric properties of the detected defects, such as depth and surface area. For measuring the width and length of the defects, we utilized segmentation masks generated during the detection phase. To address the challenge of measuring defect depth, we relied on traditional computer vision techniques applied to 3D point clouds. We reconstructed an ideal, defect-free surface from points classified as non-defective, and then determined the defect depth as the maximum distance between the defective points and this reconstructed surface.
- 1 May 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (03) , 031201-2
Special Section Guest Editorial: Quality Control by Artificial Vision VII
Guest Editors Igor Jovančević and Jean-José Orteu introduce the Special Section on Quality Control by Artificial Vision VII.
- 1 May 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (03) , 031201-2
Special Section Guest Editorial: Quality Control by Artificial Vision VII
Guest Editors Igor Jovančević and Jean-José Orteu introduce the Special Section on Quality Control by Artificial Vision VII.
- 14 February 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (03) , 031205-12
Domain shared features for visual inspection of complex mechanical assemblies based on synthetically generated data
Even though neural network methodologies have been established for a long time, only recently have they achieved exceptional efficacy in practical deployments, predominantly due to improvements in hardware computational capacity and the large amounts of available data for learning. Nonetheless, substantial challenges remain in utilizing deep learning in many domains, mainly because of the lack of large amounts of labeled data that are versatile enough for deep learning models to learn useful information. For instance, in mechanical assembly inspection, annotating data for each type of mechanical part to train a deep learning model can be very labor-intensive. Additionally, it is required to annotate data after each modification of mechanical part specification. Also, the system for inspection is typically not available until the first few samples are built to collect data. This paper proposes a solution for these challenges in case of the visual mechanical assembly inspection by processing point cloud data acquired via a three-dimensional (3D) scanner. To reduce the necessity for manually labeling large amounts of data, we employed synthetically generated data for both training and validation purposes, reserving the real sensor data exclusively for the testing phase. Our approach reduces the need for large amounts of labeled data by using synthetically generated point clouds from computer-aided design models for neural network training. Domain gap is a significant challenge for the usage of synthetically generated data. To reduce the domain gap, we used different preprocessing techniques, as well as a neural network architecture that focuses more on shared features that will not significantly change between synthetically generated data and real data from the 3D sensor.
- 14 February 2024
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 33 (03) , 031205-12
Domain shared features for visual inspection of complex mechanical assemblies based on synthetically generated data
Even though neural network methodologies have been established for a long time, only recently have they achieved exceptional efficacy in practical deployments, predominantly due to improvements in hardware computational capacity and the large amounts of available data for learning. Nonetheless, substantial challenges remain in utilizing deep learning in many domains, mainly because of the lack of large amounts of labeled data that are versatile enough for deep learning models to learn useful information. For instance, in mechanical assembly inspection, annotating data for each type of mechanical part to train a deep learning model can be very labor-intensive. Additionally, it is required to annotate data after each modification of mechanical part specification. Also, the system for inspection is typically not available until the first few samples are built to collect data. This paper proposes a solution for these challenges in case of the visual mechanical assembly inspection by processing point cloud data acquired via a three-dimensional (3D) scanner. To reduce the necessity for manually labeling large amounts of data, we employed synthetically generated data for both training and validation purposes, reserving the real sensor data exclusively for the testing phase. Our approach reduces the need for large amounts of labeled data by using synthetically generated point clouds from computer-aided design models for neural network training. Domain gap is a significant challenge for the usage of synthetically generated data. To reduce the domain gap, we used different preprocessing techniques, as well as a neural network architecture that focuses more on shared features that will not significantly change between synthetically generated data and real data from the 3D sensor.
- Open Access
- 31 January 2024
- Published by Springer Nature in Research Square
Automatic visual inspection of mechanical assemblies via 3Dpoint cloud classification with deep neural networks
In this work, we are focused on conformity control of complex aeronautical mechanical assemblies,typically an aircraft engine, at the end or in the middle of the assembly process. A 3D scannercarried by a robot arm provides acquisitions of 3D point clouds which are further processed by deepclassification neural networks. The Computer-Aided Design (CAD) model of the inspected mechanicalassembly is available, and the proposed approach relies on it. The models are trained on syntheticdata, generated from the CAD models.
- 31 January 2024
- Published by Springer Nature in Research Square
Open AccessAutomatic visual inspection of mechanical assemblies via 3Dpoint cloud classification with deep neural networks
In this work, we are focused on conformity control of complex aeronautical mechanical assemblies,typically an aircraft engine, at the end or in the middle of the assembly process. A 3D scannercarried by a robot arm provides acquisitions of 3D point clouds which are further processed by deepclassification neural networks. The Computer-Aided Design (CAD) model of the inspected mechanicalassembly is available, and the proposed approach relies on it. The models are trained on syntheticdata, generated from the CAD models.
- 20 June 2022
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 31 (05) , 051604-26
Learning deep domain-agnostic features from synthetic renders for industrial visual inspection
Deep learning has resulted in a huge advancement in computer vision. However, deep models require an enormous amount of manually annotated data, which is a laborious and time-consuming task. Large amounts of images demand the availability of target objects for acquisition. This is a kind of luxury we usually do not have in the context of automatic inspection of complex mechanical assemblies, such as in the aircraft industry. We focus on using deep convolutional neural networks (CNN) for automatic industrial inspection of mechanical assemblies, where training images are limited and hard to collect. Computer-aided design model (CAD) is a standard way to describe mechanical assemblies; for each assembly part we have a three-dimensional CAD model with the real dimensions and geometrical properties. Therefore, rendering of CAD models to generate synthetic training data is an attractive approach that comes with perfect annotations. Our ultimate goal is to obtain a deep CNN model trained on synthetic renders and deployed to recognize the presence of target objects in never-before-seen real images collected by commercial RGB cameras. Different approaches are adopted to close the domain gap between synthetic and real images. First, the domain randomization technique is applied to generate synthetic data for training. Second, domain invariant features are utilized while training, allowing to use the trained model directly in the target domain. Finally, we propose a way to learn better representative features using augmented autoencoders, getting performance close to our baseline models trained with real images.
- 20 June 2022
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 31 (05) , 051604-26
Learning deep domain-agnostic features from synthetic renders for industrial visual inspection
Deep learning has resulted in a huge advancement in computer vision. However, deep models require an enormous amount of manually annotated data, which is a laborious and time-consuming task. Large amounts of images demand the availability of target objects for acquisition. This is a kind of luxury we usually do not have in the context of automatic inspection of complex mechanical assemblies, such as in the aircraft industry. We focus on using deep convolutional neural networks (CNN) for automatic industrial inspection of mechanical assemblies, where training images are limited and hard to collect. Computer-aided design model (CAD) is a standard way to describe mechanical assemblies; for each assembly part we have a three-dimensional CAD model with the real dimensions and geometrical properties. Therefore, rendering of CAD models to generate synthetic training data is an attractive approach that comes with perfect annotations. Our ultimate goal is to obtain a deep CNN model trained on synthetic renders and deployed to recognize the presence of target objects in never-before-seen real images collected by commercial RGB cameras. Different approaches are adopted to close the domain gap between synthetic and real images. First, the domain randomization technique is applied to generate synthetic data for training. Second, domain invariant features are utilized while training, allowing to use the trained model directly in the target domain. Finally, we propose a way to learn better representative features using augmented autoencoders, getting performance close to our baseline models trained with real images.
- Open Access
- 7 April 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (8) , 3301
Learning Local Descriptor for Comparing Renders with Real Images
We present a method to train a deep-network-based feature descriptor to calculate discriminative local descriptions from renders and corresponding real images with similar geometry. We are interested in using such descriptors for automatic industrial visual inspection whereby the inspection camera has been coarsely localized with respect to a relatively large mechanical assembly and presence of certain components needs to be checked compared to the reference computer-aided design model (CAD). We aim to perform the task by comparing the real inspection image with the render of textureless 3D CAD using the learned descriptors. The descriptor was trained to capture geometric features while staying invariant to image domain. Patch pairs for training the descriptor were extracted in a semisupervised manner from a small data set of 100 pairs of real images and corresponding renders that were manually finely registered starting from a relatively coarse localization of the inspection camera. Due to the small size of the training data set, the descriptor network was initialized with weights from classification training on ImageNet. A two-step training is proposed for addressing the problem of domain adaptation. The first, “bootstrapping”, is a classification training to obtain good initial weights for second training step, triplet-loss training, that provides weights for extracting the discriminative features comparable using distance. The descriptor was tested for comparing renders and real images through two approaches: finding local correspondences between the images through nearest neighbor matching and transforming the images into Bag of Visual Words (BoVW) histograms. We observed that learning a robust cross-domain descriptor is feasible, even with a small data set, and such features might be of interest for CAD-based inspection of mechanical assemblies, and related applications such as tracking or finely registered augmented reality. To the best of our knowledge, this is the first work that reports learning local descriptors for comparing renders with real inspection images.- 7 April 2021
- Published by MDPI AG in Applied Sciences
- Vol. 11 (8) , 3301
Open AccessLearning Local Descriptor for Comparing Renders with Real Images
We present a method to train a deep-network-based feature descriptor to calculate discriminative local descriptions from renders and corresponding real images with similar geometry. We are interested in using such descriptors for automatic industrial visual inspection whereby the inspection camera has been coarsely localized with respect to a relatively large mechanical assembly and presence of certain components needs to be checked compared to the reference computer-aided design model (CAD). We aim to perform the task by comparing the real inspection image with the render of textureless 3D CAD using the learned descriptors. The descriptor was trained to capture geometric features while staying invariant to image domain. Patch pairs for training the descriptor were extracted in a semisupervised manner from a small data set of 100 pairs of real images and corresponding renders that were manually finely registered starting from a relatively coarse localization of the inspection camera. Due to the small size of the training data set, the descriptor network was initialized with weights from classification training on ImageNet. A two-step training is proposed for addressing the problem of domain adaptation. The first, “bootstrapping”, is a classification training to obtain good initial weights for second training step, triplet-loss training, that provides weights for extracting the discriminative features comparable using distance. The descriptor was tested for comparing renders and real images through two approaches: finding local correspondences between the images through nearest neighbor matching and transforming the images into Bag of Visual Words (BoVW) histograms. We observed that learning a robust cross-domain descriptor is feasible, even with a small data set, and such features might be of interest for CAD-based inspection of mechanical assemblies, and related applications such as tracking or finely registered augmented reality. To the best of our knowledge, this is the first work that reports learning local descriptors for comparing renders with real inspection images. - Open Access
- 25 April 2020
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 29 (04) , 1-22
Three-dimensional point cloud analysis for automatic inspection of complex aeronautical mechanical assemblies
The Journal of Electronic Imaging (JEI), copublished bimonthly with the Society for Imaging Science and Technology, publishes peer-reviewed papers that cover research and applications in all areas of electronic imaging science and technology.
- 25 April 2020
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 29 (04) , 1-22
Open AccessThree-dimensional point cloud analysis for automatic inspection of complex aeronautical mechanical assemblies
The Journal of Electronic Imaging (JEI), copublished bimonthly with the Society for Imaging Science and Technology, publishes peer-reviewed papers that cover research and applications in all areas of electronic imaging science and technology.
- 17 February 2020
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 29 (04) , 041008-14
Classification using a three-dimensional sensor in a structured industrial environment
Usage of a three-dimensional (3-D) sensor and point clouds provides various benefits over the usage of a traditional camera for industrial inspection. We focus on the development of a classification solution for industrial inspection purposes using point clouds as an input. The developed approach employs deep learning to classify point clouds, acquired via a 3-D sensor, the final goal being to verify the presence of certain industrial elements in the scene. We possess the computer-aided design model of the whole mechanical assembly and an in-house developed localization module provides initial pose estimation from which 3-D point clouds of the elements are inferred. The accuracy of this approach is proved to be acceptable for industrial usage. Robustness of the classification module in relation to the accuracy of the localization algorithm is also estimated.
- 17 February 2020
- Published by SPIE-Intl Soc Optical Eng in Journal of Electronic Imaging
- Vol. 29 (04) , 041008-14
Classification using a three-dimensional sensor in a structured industrial environment
Usage of a three-dimensional (3-D) sensor and point clouds provides various benefits over the usage of a traditional camera for industrial inspection. We focus on the development of a classification solution for industrial inspection purposes using point clouds as an input. The developed approach employs deep learning to classify point clouds, acquired via a 3-D sensor, the final goal being to verify the presence of certain industrial elements in the scene. We possess the computer-aided design model of the whole mechanical assembly and an in-house developed localization module provides initial pose estimation from which 3-D point clouds of the elements are inferred. The accuracy of this approach is proved to be acceptable for industrial usage. Robustness of the classification module in relation to the accuracy of the localization algorithm is also estimated.