Article
Computer Science, Artificial Intelligence
Xiaobin Hu, Wenqi Ren, Jiaolong Yang, Xiaochun Cao, David Wipf, Bjoern Menze, Xin Tong, Hongbin Zha
Summary: This paper proposes an improved face restoration method by embedding the network with 3D morphable priors, which enhances the performance of facial restoration tasks. Experimental results demonstrate superior performance of this method in face super-resolution and deblurring.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Chemistry, Multidisciplinary
Awais Salman Qazi, Muhammad Shoaib Farooq, Furqan Rustam, Monica Gracia Villar, Carmen Lili Rodriguez, Imran Ashraf
Summary: Facial emotion recognition is an important research topic in pattern recognition, with applications in various fields. This paper proposes a new architecture of a convolutional neural network for the FER system, which achieves higher accuracy through feature map enhancement. An application is developed to identify basic human expressions.
APPLIED SCIENCES-BASEL
(2022)
Article
Health Care Sciences & Services
Ziwei Song, Kristie Nguyen, Tien Nguyen, Catherine Cho, Jerry Gao
Summary: Wearing a face mask is important for preventing the spread of COVID-19, but it poses challenges for facial recognition technology. The Spartan Face Detection and Facial Recognition System proposes a solution using deep learning algorithms to detect masks, classify their type and position, and recognize individuals with masks.
Article
Computer Science, Artificial Intelligence
Zhengzheng Sun, Lianfang Tian, Qiliang Du, Jameel A. Bhutto, Zhaolin Wang
Summary: Face Super-Resolution (FSR) is an important research topic in image restoration field. This study proposes a Facial Mask Attention Network based on deep convolutional neural networks to generate high-resolution and identity-faithful face images. By using a selective pixel loss function, the model emphasizes on face regions with dense identity features, resulting in more natural and detailed restored face images compared to state-of-the-art methods.
NEURAL COMPUTING & APPLICATIONS
(2023)
Article
Computer Science, Information Systems
Naeem Ullah, Ali Javed, Mustansar Ali Ghazanfar, Abdulmajeed Alsufyani, Sami Bourouis
Summary: The COVID-19 pandemic has greatly impacted people's daily life worldwide, prompting the recommendation of wearing face masks in public places. However, manual inspection of mask-wearing and traditional face recognition techniques are ineffective in this context. This paper proposes a novel framework called DeepMasknet that can detect face masks and recognize individuals wearing masks. Additionally, a large-scale and diverse dataset called MDMFR was developed for evaluating the performance of both mask detection and masked facial recognition methods.
JOURNAL OF KING SAUD UNIVERSITY-COMPUTER AND INFORMATION SCIENCES
(2022)
Article
Computer Science, Information Systems
Muhammad Haikal Mohd Kamil, Norliza Zaini, Lucyantie Mazalan, Afiq Harith Ahamad
Summary: This paper presents an online attendance system that uses facial recognition and facial mask detection. The objective is to develop an effective attendance system based on these technologies and make it accessible online through a browser interface. The system allows users to easily record attendance data online and stores it in a centralized database.
MULTIMEDIA TOOLS AND APPLICATIONS
(2023)
Article
Chemistry, Analytical
Caijie Zhao, Ying Qin, Bob Zhang
Summary: The ALOB model, which mitigates the need for manual labelling of occlusions, achieves excellent results in occluded face recognition by contrastively learning corrupted features against personal identity labels. It outperforms existing state-of-the-art methods in various experiments and demonstrates superior performance in masked face recognition and general face recognition.
Article
Computer Science, Information Systems
Xingyu Yang, Mengya Han, Yong Luo, Han Hu, Yonggang Wen
Summary: This paper proposes a novel Two-stream Prototype Learning Network (TSPLN) for few-shot face recognition under occlusion. The method addresses the challenge of deteriorated class prototypes due to unknown occlusions, and focuses on the quality of support images and their relevance to the query image for recognition improvement. Extensive experiments demonstrate that the proposed method achieves state-of-the-art performance for occluded face recognition in the few-shot setting.
IEEE TRANSACTIONS ON MULTIMEDIA
(2023)
Article
Computer Science, Artificial Intelligence
Xiangyu Zhu, Hongyan Fei, Bin Zhang, Tianshuo Zhang, Xiaoyu Zhang, Stan Z. Li, Zhen Lei
Summary: Detecting digital face manipulation has been a focus of attention due to the risks posed by fake media. This study proposes a novel 3D decomposition based method that disentangles a face image into components such as 3D shape, lighting, common texture, and identity texture. It also introduces a fine-grained morphing network and a composition search strategy to extract forgery clues. Experimental results show that the decomposed components highlight forgery artifacts and the searched architecture extracts discriminative forgery features, achieving state-of-the-art performance.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2023)
Article
Computer Science, Artificial Intelligence
Tao Lu, Yuanzhi Wang, Yanduo Zhang, Junjun Jiang, Zhongyuan Wang, Zixiang Xiong
Summary: This article proposes a novel pre-prior guided approach that extracts facial prior information from high-resolution face images and embeds them into low-resolution ones, resulting in high-frequency information-rich low-resolution face images and improved face reconstruction performance.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Artificial Intelligence
Yiming Wang, Xinghui Dong, Gongfa Li, Junyu Dong, Hui Yu
Summary: Facial expression recognition has made significant progress, but recognizing expressions with occlusions and large head-poses remains challenging. This paper proposes a cascade regression-based face frontalization method, which improves the performance of static and dynamic facial expression recognition by predicting frontal shape and reconstructing facial texture.
COGNITIVE COMPUTATION
(2022)
Article
Computer Science, Information Systems
Andre Sobiecki, Julius van Dijk, Hidde Folkertsma, Alexandru Telea
Summary: The study found that face verification methods perform well on standardized face images, but face challenges on low resolution, poor lighting, and non-standard face positions. Less than half of face restoration methods help with face verification, with some methods with lower quality evaluations actually being the most helpful. Experiments show that face verification works less effectively as resolution decreases.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Multidisciplinary Sciences
Jordan R. Saadon, Fan Yang, Ryan Burgert, Selma Mohammad, Theresa Gammel, Michael Sepe, Miriam Rafailovich, Charles B. Mikell, Pawel Polak, Sima Mofakham
Summary: Research into mood and emotion has often relied on slow and subjective self-report, and there is a need for rapid, accurate, and objective assessment tools. In this study, a method using digital image speckle correlation (DISC) was developed to track subtle changes in facial expressions and assess emotions in real-time. The results showed that DISC-based classifiers can reliably identify an individual's emotion with better predictions compared to existing commercial solutions, and they are free of racial or gender bias.
Article
Surgery
Miguel I. Dorante, Alice T. Wang, Branislav Kollar, Bridget J. Perry, Mustafa G. Ertosun, Andrew J. Lindford, Emma-Lotta Kiukas, Oemer Oezkan, Oezlenen Oezkan, Patrik Lassus, Bohdan Pomahac
Summary: Software-based analysis with FaceReader can be used to assess motor function restoration after face transplant, and international collaboration strengthens the reliability of outcome data.
PLASTIC AND RECONSTRUCTIVE SURGERY
(2023)
Article
Computer Science, Information Systems
Delphine Poux, Benjamin Allaert, Jose Mennesson, Nacim Ihaddadene, Ioan Marius Bilasco, Chaabane Djeraba
Summary: The research focuses on recognizing facial expressions in the presence of occlusions by exploiting the specificities of facial movement propagation. It constructs adapted facial frameworks for each expression, computes the importance of unoccluded facial regions, and aggregates the outputs of expression-dependant binary classifiers to build a unique model for recognizing all considered facial expressions. The evaluations demonstrate the robustness of this approach in handling significant facial occlusions.
MULTIMEDIA TOOLS AND APPLICATIONS
(2021)
Article
Computer Science, Information Systems
Luca Ulrich, Enrico Vezzetti, Sandro Moos, Federica Marcolin
MULTIMEDIA TOOLS AND APPLICATIONS
(2020)
Article
Biochemistry & Molecular Biology
Luca Di Grazia, Maral Aminpour, Enrico Vezzetti, Vahid Rezania, Federica Marcolin, Jack Adam Tuszynski
Summary: This study successfully translated biometric 3D face recognition concepts and algorithms into protein biophysics, achieving rapid and precise classification of protein surface morphological features. The results indicate a potential application of this method in protein structure analysis, showing significant competitiveness with existing protein classification methods.
PROTEINS-STRUCTURE FUNCTION AND BIOINFORMATICS
(2021)
Article
Dentistry, Oral Surgery & Medicine
Giovanni Gerbino, Umberto Autorino, Claudia Borbon, Federica Marcolin, Elena Olivetti, Enrico Vezzetti, Emanuele Zavattero
Summary: This prospective study evaluated the 3D soft tissue and bone changes in the malar region of 10 patients undergoing malar valgization osteotomy in conjunction with orthognatic surgery. Results showed improvements in zygomatic projection and balanced facial correction, with an increase in facial volume and surface analysis. The surgery was performed safely and effectively, contributing to overall patient satisfaction and aesthetic outcomes.
JOURNAL OF CRANIO-MAXILLOFACIAL SURGERY
(2021)
Article
Health Care Sciences & Services
Elena Carlotta Olivetti, Federica Marcolin, Sandro Moos, Alberto Ferrando, Enrico Vezzetti, Umberto Autorino, Claudia Borbon, Emanuele Zavattero, Giovanni Gerbino, Guglielmo Ramieri
Summary: This study introduces a novel 3D approach for evaluating soft tissue changes after zygomatic osteotomy, using geometrical descriptors to detect modifications in facial surface shape, showing a higher sensitivity compared to angular evaluation.
JOURNAL OF PERSONALIZED MEDICINE
(2021)
Article
Computer Science, Artificial Intelligence
F. Marcolin, E. Vezzetti, M. G. Monaci
Summary: Human beings are still the most effective face recognition system, able to detect familiar faces in various conditions. Therefore, neuroscientists and psychologists are studying the human visual system and face perception techniques to understand the processes behind human vision.
Article
Chemistry, Multidisciplinary
Ivonne Angelica Castiblanco Jimenez, Laura Cristina Cepeda Garcia, Federica Marcolin, Maria Grazia Violante, Enrico Vezzetti
Summary: Supporting education and training initiatives is an effective way to address Sustainable Development Challenges, with e-learning being a viable option due to its advantages. Understanding the reasons for user adoption of such technologies is crucial for achieving social objectives. The study validates a TAM extension for e-learning in agriculture, finding that content quality significantly influences usefulness perception, while experience and self-efficacy impact ease of use.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Software Engineering
Federica Marcolin, Giulia Wally Scurati, Luca Ulrich, Francesca Nonis, Enrico Vezzetti, Nicolo Dozio, Francesco Ferrise
Summary: Virtual Reality goes beyond visual representation to stimulate all human senses and allow users to interact with virtual environments, providing experiences comparable to real-life ones. With the potential to evoke various emotions, VR technology offers a new way to explore and impact human emotions and improve quality of life.
IEEE COMPUTER GRAPHICS AND APPLICATIONS
(2021)
Article
Multidisciplinary Sciences
Nicolo Dozio, Federica Marcolin, Giulia Wally Scurati, Francesca Nonis, Luca Ulrich, Enrico Vezzetti, Francesco Ferrise
Summary: This paper describes the validation of ten affective interactive Virtual Environments (VEs) for use in Virtual Reality, showing that scenarios can be differentiated based on the emotion aroused. Results demonstrate high reliability and strong adaptability of the experiences to different contexts of use.
SCIENTIFIC REPORTS
(2021)
Article
Chemistry, Multidisciplinary
Luca Ulrich, Francesca Nonis, Enrico Vezzetti, Sandro Moos, Giandomenico Caruso, Yuan Shi, Federica Marcolin
Summary: The study experimented the impact of ADAS systems on attention and accidents, using deep learning for facial expression recognition. It found a correlation between attention and accidents, but no relationship between facial expressions and ADAS activations.
APPLIED SCIENCES-BASEL
(2021)
Article
Computer Science, Information Systems
Ivonne Angelica Castiblanco Jimenez, Stefano Mauro, Domenico Napoli, Federica Marcolin, Enrico Vezzetti, Maria Camila Rojas Torres, Stefania Specchia, Sandro Moos
Summary: The development of new waste disposal methods is crucial for eco-friendly cities, with waste management being no exception. This research explores a more ecological and sustainable solution for infectious waste treatment in Northern Italy, aiming to improve the population's quality of life. By utilizing Design Thinking, feasible results based on user needs can be obtained, contributing to the creation of better cities for all.
Article
Computer Science, Cybernetics
Nicolo Dozio, Federica Marcolin, Giulia Wally Scurati, Luca Ulrich, Francesca Nonis, Enrico Vezzetti, Gabriele Marsocci, Alba La Rosa, Francesco Ferrise
Summary: This paper discusses how to combine multiple design elements to elicit five distinct emotions and presents the methodology, development of case studies, and testing results.
INTERNATIONAL JOURNAL OF HUMAN-COMPUTER STUDIES
(2022)
Article
Computer Science, Artificial Intelligence
Giulio Mangano, Andrea Ferrari, Carlo Rafele, Enrico Vezzetti, Federica Marcolin
Summary: The ongoing research on facial expression recognition has been driven by advancements in deep learning algorithms, making it a part of the Artificial Intelligence literature. Although its applications in areas like human-computer interaction and marketing are emerging, only a few studies have explored the willingness of end users to share their facial data for emotion detection. This study investigates the awareness and interest levels of 373 potential consumers in the car insurance sector, focusing on differentiating between Generation Y and Z respondents. Results indicate that younger individuals, those with higher education levels, and social media users are more inclined to share their expressive facial data.
Article
Computer Science, Information Systems
Ivonne Angelica Castiblanco Jimenez, Francesca Nonis, Elena Carlotta Olivetti, Luca Ulrich, Sandro Moos, Maria Grazia Monaci, Federica Marcolin, Enrico Vezzetti
Summary: In recent years, museums and exhibitions have utilized advancements in Virtual Reality technologies to enhance visitors' engagement, interaction, comprehension, and accessibility to collections. Research suggests that self-assessment techniques may not fully assess visitors' real affective state, prompting the adoption of physiological techniques like EEG for a more unbiased understanding of their feelings. This study analyzes the affective state of 95 visitors interacting with handicraft objects physically or virtually using EEG-based indicators. The results highlight the potential of VR technologies in enhancing participants' cognitive engagement and suggest the viability of using EEG-based indicators as a "ground truth of emotion".
Article
Education & Educational Research
Andrea Catalina Ladino Nocua, Joan Paola Cruz Gonzalez, Ivonne Angelica Castiblanco Jimenez, Juan Sebastian Gomez Acevedo, Federica Marcolin, Enrico Vezzetti
Summary: Student engagement is crucial for educational institutions to improve teaching methods and evaluate the quality of education, especially in the context of distance learning during the COVID-19 pandemic. Monitoring cognitive engagement through physiological parameters like heart rate can provide valuable insights. Additionally, informing students in advance about upcoming tasks can significantly increase their cognitive engagement levels.
EDUCATION SCIENCES
(2021)
Article
Computer Science, Information Systems
Ivonne Angelica Castiblanco Jimenez, Laura Cristina Cepeda Garcia, Maria Grazia Violante, Federica Marcolin, Enrico Vezzetti
Summary: In recent years, ICT has had a significant impact on various sectors of modern society, including agriculture. This study identifies common external variables for further validation in e-learning tools designed for EU farmers and agricultural entrepreneurs. The research also shows the different impacts of external variables on the main beliefs of the TAM model, Perceived Usefulness and Perceived Ease of Use.
