Article
Mathematics, Applied
Chengzhuan Yang, Qian Yu
Summary: Shape is an important visual characteristic that represents objects, and shape recognition is a significant research direction. The invariant multiscale triangle feature (IMTF) is a novel method for robust shape recognition, which can effectively combine boundary and interior characteristics to improve recognition accuracy.
APPLIED MATHEMATICS AND COMPUTATION
(2021)
Article
Computer Science, Artificial Intelligence
Mohammed Ayoub Alaoui Mhamdi, Djemel Ziou
Summary: This paper presents a critical points based descriptor for 3 D objects recognition, utilizing a size function to represent critical points and the links between them, and a metric learning method to deal with partial matching problems. Experimental results show that the proposed method performs well in 3 D object recognition.
PATTERN RECOGNITION
(2021)
Article
Computer Science, Artificial Intelligence
Michalis Lazarou, Bo Li, Tania Stathaki
Summary: This work introduces a novel shape matching methodology for real-time hand gesture recognition and demonstrates its superiority in accuracy and computational efficiency through comparison with other methods.
COMPUTER VISION AND IMAGE UNDERSTANDING
(2021)
Review
Chemistry, Analytical
Jose Maria Martinez-Otzeta, Itsaso Rodriguez-Moreno, Inigo Mendialdua, Basilio Sierra
Summary: Random Sample Consensus (RANSAC) is a robust estimation method for models contaminated by outliers. It starts with sample selection, evaluates the adequacy of the estimation, and repeats the process until a stopping criterion is met. RANSAC is widely used in robotics, particularly for finding geometric shapes in point clouds or estimating camera view transformations.
Article
Computer Science, Artificial Intelligence
Jiaqi Yang, Ke Xian, Peng Wang, Yanning Zhang
Summary: This paper provides a comprehensive evaluation of nine state-of-the-art 3D correspondence grouping methods under various application contexts and perturbations, aiming to find a more efficient and accurate way for point-to-point correspondences between 3D rigid data.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2021)
Article
Multidisciplinary Sciences
Jason N. Bruck, Sam F. Walmsley, Vincent M. Janik
Summary: This study found that dolphins can distinguish individuals through gustatory stimuli and integrate information from acoustic and taste inputs, indicating the existence of a modality independent concept for known conspecifics.
Article
Computer Science, Artificial Intelligence
Zhihong Sun, Jun Chen, Mithun Mukherjee, Chao Liang, Weijian Ruan, Zhigeng Pan
Summary: This paper proposes a new method that combines global and partial feature matching models to improve similarity measurement between targets, and introduces a detection modifier method based on human pose information. Experimental results demonstrate that the method achieves comparable performance in multi-object tracking tasks.
Article
Multidisciplinary Sciences
Tijl Grootswagers, Ivy Zhou, Amanda K. Robinson, Martin N. Hebart, Thomas A. Carlson
Summary: This paper presents the THINGS-EEG dataset, which includes electroencephalography responses from 50 subjects to 1,854 object concepts and 22,248 images. This dataset can support research in understanding how the brain recognizes and processes visual objects.
Article
Neurosciences
Anna Bognar, Rufin Vogels
Summary: Current models of object recognition are based on spatial representations formed from object features in retinal images. Anorthoscopic perception, which involves recognizing objects behind occlusions or through narrow slits, requires spatiotemporal integration of shape parts. Studies on monkey IT neurons showed that while the neurons signal shape identity during slit-viewing, spatiotemporal integration for whole shape perception may occur downstream to IT.
JOURNAL OF NEUROSCIENCE
(2021)
Article
Computer Science, Artificial Intelligence
Yong Xu, Chaoda Zheng, Ruotao Xu, Yuhui Quan, Haibin Ling
Summary: In recent years, multi-view learning has become a promising approach for 3D shape recognition by identifying shapes based on 2D views from different angles. This paper proposes a correspondence-aware representation (CAR) module that finds potential intra-view and cross-view correspondences through kNN search in semantic space and aggregates shape features via learned transforms. Incorporating the CAR module into a ResNet-18 backbone, an effective deep model called CAR-Net is introduced for 3D shape classification and retrieval, demonstrating the effectiveness and excellent performance of the CAR module.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2021)
Article
Geochemistry & Geophysics
Wuyong Tao, Xianghong Hua, Kegen Yu, Xijiang Chen, Bufan Zhao
Summary: This article presents an object recognition pipeline using a highly descriptive, robust, and computationally efficient local shape descriptor (LSD) to establish correspondences, a clustering method utilizing local reference frames (LRFs) of keypoints to select correct correspondences, and an index to verify transformation hypotheses. Experimental results show high descriptor matching performance, effective grouping of correct correspondences, and efficient filtering of false transformation hypotheses, enhancing recognition performance.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Computer Science, Artificial Intelligence
Yashil Sukurdeep, Martin Bauer, Nicolas Charon
Summary: This paper introduces a new method for shape registration and matching of shape graphs with topological inconsistencies. By using higher-order invariant Sobolev metrics and varifolds, we are able to tackle the matching problem between shape graphs and obtain solutions through numerical optimization approaches.
SIAM JOURNAL ON IMAGING SCIENCES
(2022)
Article
Computer Science, Theory & Methods
Mandi Luo, Xin Ma, Zhihang Li, Jie Cao, Ran He
Summary: The proposed method introduces a Saliency Search Network (SSN) to extract domain-invariant identity features for NIR-VIS HFR, addressing the challenges posed by modality gaps and occlusions. By automatically searching for efficient parts of face images and guided by an information bottleneck network, it effectively deals with modality discrepancy and occlusions.
IEEE TRANSACTIONS ON INFORMATION FORENSICS AND SECURITY
(2021)
Article
Computer Science, Information Systems
Zahra Hossein-Nejad, Mehdi Nasri
Summary: This paper proposes a new approach for object recognition in remote-sensing images, using Scale Invariant Feature Transform (SIFT) for matching the object in the template and test images. An adaptive Random sample consensus (RANSAC) algorithm is used to reduce false matches of SIFT, and the extended region-growing algorithm is used to extract the exact object boundary.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Automation & Control Systems
Seungyeon Kim, Taegyun Ahn, Yonghyeon Lee, Jihwan Kim, Michael Yu Wang, Frank C. Park
Summary: This paper proposes a two-step method that combines deformable superquadrics with a deep learning network to identify and grasp partially occluded objects. Experimental results show improved success rates and faster recognition compared to existing methods.
IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING
(2023)
Review
Environmental Sciences
Jacopo Aguzzi, Sascha Flogel, Simone Marini, Laurenz Thomsen, Jan Albiez, Peter Weiss, Giacomo Picardi, Marcello Calisti, Sergio Stefanni, Luca Mirimin, Fabrizio Vecchi, Cecilia Laschi, Andrew Branch, Evan B. Clark, Bernard Foing, Armin Wedler, Damianos Chatzievangelou, Michael Tangherlini, Autun Purser, Lewis Dartnell, Roberto Danovaro
Summary: Recent advances in robotic design, autonomy and sensor integration are enabling exploration of deep-sea environments and can be applied to oceans of icy moons. Synergies with space technologies hold promise for the future development of deep-sea robotics in areas such as biomimetic structure and propulsion, artificial intelligence and cooperative networks, and life-detecting instrument design.
ELEMENTA-SCIENCE OF THE ANTHROPOCENE
(2022)
Article
Environmental Sciences
Sergio Stefanni, Luca Mirimin, David Stankovic, Damianos Chatzievangelou, Lucia Bongiorni, Simone Marini, Maria Vittoria Modica, Elisabetta Manea, Federico Bonofiglio, Joaquin del Rio Fernandez, Neven Cukrov, Ana Gavrilovic, Fabio C. De Leo, Jacopo Aguzzi
Summary: Deep-sea ecosystems are reservoirs of largely unexplored biodiversity, and the application of environmental DNA (eDNA) technology is revolutionizing the monitoring of deep-sea biodiversity. This article describes the potential of eDNA as a core component for ecological monitoring capabilities and its integration with optoacoustic imaging and other multidisciplinary data. It also discusses the cross-linking of eDNA with other biodiversity databases, the use of artificial intelligence in eDNA analysis, and the benefits of eDNA augmented observatories for the conservation and sustainable management of deep-sea biodiversity.
FRONTIERS IN MARINE SCIENCE
(2022)
Article
Ecology
Simone Marini, Federico Bonofiglio, Lorenzo P. Corgnati, Andrea Bordone, Stefano Schiaparelli, Andrea Peirano
Summary: The rapid climate changes in Antarctica present challenges to living communities and call for urgent monitoring efforts. This study presents a pilot project utilizing automated cameras and image processing techniques to establish a non-invasive and sustainable monitoring activity. The results demonstrate the effectiveness of autonomous imaging devices and image analysis algorithms in acquiring and extracting relevant scientific knowledge from long-term visual data. The success of this study represents a significant step towards continuous data collection in Antarctic and remote underwater habitats.
METHODS IN ECOLOGY AND EVOLUTION
(2022)
Article
Environmental Sciences
Ennio Ottaviani, Marco Francescangeli, Nikolla Gjeci, Joaquin del Rio Fernandez, Jacopo Aguzzi, Simone Marini
Summary: The marine science community is focused on exploring and monitoring biodiversity dynamics, particularly in understanding ecosystem functioning and tracking human impacts. Innovative technological solutions are needed for accurate monitoring of marine ecosystems with remote and continuous data collection. Automated intelligent services are required to extract relevant biological information from visual data, but concept drift affects the performance of automated detection and classification.
FRONTIERS IN MARINE SCIENCE
(2022)
Editorial Material
Engineering, Marine
Fausto Pedro Garcia Marquez, Mayorkinos Papaelias, Simone Marini
JOURNAL OF MARINE SCIENCE AND ENGINEERING
(2022)
Article
Computer Science, Information Systems
Andreas Scalas, Daniela Cabiddu, Michela Mortara, Michela Spagnuolo
Summary: An urban digital twin is a virtual representation of a city's real assets, processes, systems, and subsystems, which learns and evolves with the physical city using heterogeneous data. This paper focuses on the geometric layer of the city digital twin, discussing the challenges and potential related to it. The authors describe their approach to developing the geometric layer using pre-existing public data from aerial surveys and OpenStreetMap, and demonstrate its potential through two geometric characterization services.
ISPRS INTERNATIONAL JOURNAL OF GEO-INFORMATION
(2022)
Review
Environmental Sciences
Jacopo Aguzzi, Damianos Chatzievangelou, Nathan J. Robinson, Nixon Bahamon, Alan Berry, Marc Carreras, Joan Batista Company, Corrado Costa, Joaquin del Rio Fernandez, Ahmad Falahzadeh, Spyros Fifas, Sascha Floegel, Jordi Grinyo, Jonas Pall Jonasson, Patrik Jonsson, Colm Lordan, Mathieu Lundy, Simone Marini, Michela Martinelli, Ivan Masmitja, Luca Mirimin, Atif Naseer, Joan Navarro, Narcis Palomeras, Giacomo Picardi, Cristina Silva, Sergio Stefanni, Maria Vigo, Yolanda Vila, Adrian Weetman, Jennifer Doyle
Summary: The article introduces how innovative ecological monitoring technologies can be used to improve the accuracy of stock assessments for the Norway lobster. By using robotic platforms, telemetry, environmental DNA, and Artificial Intelligence tools, it is possible to better observe and count the lobsters and their burrow systems, as well as provide important insights into their burrowing behavior.
FRONTIERS IN MARINE SCIENCE
(2022)
Correction
Chemistry, Analytical
Vanesa Lopez-Vazquez, Jose Manuel Lopez-Guede, Simone Marini, Emanuela Fanelli, Espen Johnsen, Jacopo Aguzzi
Article
Chemistry, Multidisciplinary
Martina Paccini, Giuseppe Patane, Michela Spagnuolo
Summary: This work focuses on characterizing the morphology and pathologies of patient-specific muscle-skeletal districts to support diagnostic activities and follow-up exams. Different methods are proposed to integrate morphological information from 3D surface models with tissue information from volume images. The approach allows for qualitative and quantitative validation of bone erosion sites and the identification of osteoporotic fractures, supporting surgery planning and early diagnosis.
APPLIED SCIENCES-BASEL
(2022)
Article
Multidisciplinary Sciences
Marco Francescangeli, Simone Marini, Enoc Martinez, Joaquin Del Rio, Daniel M. Toma, Marc Nogueras, Jacopo Aguzzi
Summary: Multiparametric video-cabled marine observatories are being used to remotely monitor the marine ecosystem in real-time. These platforms provide continuous, high-frequency, and long-lasting image data sets that require automation for extracting biological time series. The OBSEA, located at a depth of 20 meters and 4 km from Vilanova i la Geltru, produced coastal fish time series continuously for 24 hours during 2013-2014 and resulted in the identification of 69,917 fish tags belonging to 30 taxa through image tagging. This tagged fish dataset is valuable for developing Artificial Intelligence routines for automated identification and classification of fishes in extensive time-lapse image sets.
Article
Environmental Sciences
Andrea Peirano, Andrea Bordone, Lorenzo P. Corgnati, Simone Marini
Summary: One-year time-lapse images were analyzed to study the abundance and activity of Odontaster validus and Sterechinus neumayeri on the sponge Mycale (Oxymycale) acerata in Tethys Bay, Antarctica. The study found that sea urchins were more abundant on the rocky bottom and sponge during different seasons, while sea stars showed a decrease in numbers on the sponge throughout the year. Both species showed peak grazing activity from January to April, coinciding with a phytoplankton bloom. The winter months were critical for both species, with the sea urchin reducing its activity.
Article
Computer Science, Software Engineering
T. Sorgente, S. Biasotti, G. Manzini, M. Spagnuolo
Summary: We analyze the efforts of the geometry processing and numerical analysis communities to define and measure mesh quality. Researchers aim to determine the impact of the mesh on the accuracy of a numerical simulation or scientific computation, and identify the geometric features that influence the results. We discuss common quality indicators and algorithms for mesh optimization.
COMPUTER GRAPHICS FORUM
(2023)
Proceedings Paper
Automation & Control Systems
Pedro Jose Bernalte Sanchez, Fausto Pedro Garcia Marquez, Mayorkinos Papaelias, Simone Marini, Nikolla Gjeci
Summary: The use of autonomous vehicles in marine and submarine works has greatly advanced in the past decade, thanks to new imaging, navigation, and communications technologies. Autonomous Underwater Vehicles (AUVs) are currently employed in various offshore missions and applications, including innovative purposes related to sustainable development and green energy mobility. The ENDURUNS project, an European research initiative, aims to conduct seabed surveys using a hydrogen fuel cell-powered underwater vehicle to achieve zero emissions. This paper analyzes the environmental management of the product using the Life Cycle Assessment methodology (ISO 14040), with the Eco-Indicator 99 method and SimaPro software for evaluating the life cycle environmental impact.
IOT AND DATA SCIENCE IN ENGINEERING MANAGEMENT
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Pedro Jose Bernalte Sanchez, Fausto Pedro Garcia Marquez, Mayorkinos Papaelias, Simone Marini, Shashank Govindaraj, Lilian Durand
Summary: In recent years, there has been an increased interest in the exploitation of offshore areas and significant growth in the marine industry. This has led to the use of autonomous marine vehicles for monitoring, surveying, and maintenance works. The ENDURUNS project, a European initiative, aims to develop a sustainable offshore exploration system innovation based on renewable energies. The project faces technical challenges in achieving zero emission performance, and the development of communication infrastructures is crucial for real-time monitoring of the system by the user.
UBIQUITOUS INTELLIGENT SYSTEMS
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Martina Paccini, Giuseppe Patane, Michela Spagnuolo
Summary: Efficient medical evaluation and monitoring of pathology progression are crucial for patients with rheumatic diseases. This paper proposes a geometry-based and texture-based approach to localize bone erosion sites, a typical symptom of rheumatic disease progression, and provides a more complete tool for analysis and visualization.
IMAGE ANALYSIS AND PROCESSING, ICIAP 2022 WORKSHOPS, PT I
(2022)