Article
Automation & Control Systems
Shuwen Zhao, Xinming Wang, Dinghuang Zhang, Gongyue Zhang, Zhiyong Wang, Honghai Liu
Summary: In this article, a novel framework is proposed to learn personalized shapes for 3D face reconstruction. Several principles are applied to balance the facial shape and expression distribution, and a mesh editing method is used to generate face images with various expressions. The pose estimation accuracy is improved by transferring the projection parameter into Euler angles, and a weighted sampling method is proposed to improve the robustness of the training process. Experimental results demonstrate that our method achieves state-of-the-art performance.
IEEE TRANSACTIONS ON CYBERNETICS
(2023)
Article
Computer Science, Artificial Intelligence
Shangfei Wang, Yanan Chang, Can Wang
Summary: This article introduces a novel multi-task dual learning framework for simultaneously addressing facial landmark detection and action unit recognition tasks and exploiting their relationship. By sharing middle-level features, common patterns can be used for facial landmark detection and AU recognition, using middle- and high-level features respectively. Furthermore, a dual learning mechanism is designed to convert predicted landmarks and AUs to corresponding facial images, exploring the strong correlations between the tasks. Jointly training the proposed method at both the feature and label levels improves each task. Experiments on two benchmark databases demonstrate the proposed method's ability to leverage dependencies and boost the generalization of both tasks.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Jun Wan, Jun Liu, Jie Zhou, Zhihui Lai, Linlin Shen, Hang Sun, Ping Xiong, Wenwen Min
Summary: Facial landmark detection methods often predict landmarks by mapping input facial appearance features to landmark heatmaps, but they struggle with large poses, occlusions, and varied illuminations. To address this, we propose a novel Reference Heatmap Transformer (RHT) that utilizes reference heatmap information for more accurate detection of facial landmarks. Experimental results on benchmark datasets show that our method outperforms existing state-of-the-art techniques.
IEEE TRANSACTIONS ON IMAGE PROCESSING
(2023)
Article
Computer Science, Information Systems
Shinfeng D. Lin, Paulo E. Linares Otoya
Summary: This paper presents a novel approach to achieve pose-invariant face recognition using ensemble learning and local feature descriptors. The proposed method extracts feature vectors from image regions surrounding specific facial landmarks, and trains three different classification models as base learners. Experimental results show better performance compared to existing methods using the CMU-PIE dataset.
Article
Computer Science, Artificial Intelligence
Jun Wan, Zhihui Lai, Jing Li, Jie Zhou, Can Gao
Summary: This article proposes a multiorder multiconstraint deep network (MMDN) for learning more powerful feature correlations and shape constraints, including an implicit multiorder correlating geometry-aware (IMCG) model and an explicit probability-based boundary-adaptive regression (EPBR) method. Experimental results demonstrate the superior performance of MMDN on challenging benchmark data sets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2022)
Article
Computer Science, Information Systems
Domenick D. Poster, Shuowen Hu, Nathan J. Short, Benjamin S. Riggan, Nasser M. Nasrabadi
Summary: This study focuses on face landmark detection in the thermal infrared spectrum, highlighting the challenges posed by differences in spatial resolution and labeling availability compared to the visible spectrum. The authors propose a visible-to-thermal parameter transfer learning method which utilizes visible face data to improve thermal face landmark detection performance. Experimental results demonstrate that transfer learning approaches such as LLR, LKR, and RPT outperform baseline models and Active Appearance Models trained solely on thermal data, showing the effectiveness of leveraging visible spectrum data in thermal face landmark detection.
Article
Geochemistry & Geophysics
Zhibo Rao, Mingyi He, Zhidong Zhu, Yuchao Dai, Renjie He
Summary: This study introduces a novel multitask learning architecture BGA-Net that can simultaneously handle semantic segmentation and disparity estimation tasks, showing notable superiority and flexibility in experiments.
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2021)
Article
Computer Science, Artificial Intelligence
Pengcheng Gao, Ke Lu, Jian Xue, Jiayi Lyu, Ling Shao
Summary: This article proposes a lightweight and efficient facial landmark detection model called EfficientFAN, which adopts the encoder-decoder structure and extracts deep dark knowledge through feature-aligned distillation and patch similarity distillation. Experimental results demonstrate the superiority of EfficientFAN over state-of-the-art methods on public datasets.
IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS
(2023)
Article
Engineering, Electrical & Electronic
Dong Liu, Long Wang, Yu Du, Ming Cong, Yongyao Li
Summary: In this study, a real-time and accurate automatic detection and segmentation algorithm for 3-D MR and TRUS images of the prostate is proposed, which achieves efficient and accurate detection and segmentation even under poor image quality conditions. The experimental results on public and private datasets demonstrate the superiority of this method.
IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT
(2022)
Article
Computer Science, Artificial Intelligence
Romain Belmonte, Benjamin Allaert, Pierre Tirilly, Ioan Marius Bilasco, Chaabane Djeraba, Nicu Sebe
Summary: In this paper, the impact of facial landmark localization (FLL) approaches on facial expression recognition (FER) tasks is studied. Due to the limitations of existing FLL datasets in measuring performance under different difficulties, the performance of recent approaches is quantified under variations in head pose and facial expressions. The study shows that optimizing the euclidean distance for landmark accuracy does not necessarily improve FER performance. To address this issue, a new evaluation metric for FLL that is more relevant to FER is proposed.
IEEE TRANSACTIONS ON AFFECTIVE COMPUTING
(2023)
Article
Computer Science, Artificial Intelligence
Jingyan Fan, Jiuzhen Liang, Hao Liu, Zhan Huan, Zhenjie Hou, Xinwen Zhou
Summary: This paper proposes a probability-guided hourglass network to enhance the shape constraints for robust facial landmark detection in unconstrained environments. The method involves a multi-scale pre-processing module for feature extraction, generating coarse localizations and probability maps based on heatmaps, and a probability-based boundary regression method with modified hausdorff distance as the loss function. The experimental results demonstrate that this method outperforms state-of-the-arts on unconstrained conditions.
IET IMAGE PROCESSING
(2023)
Article
Automation & Control Systems
Van-Thanh Hoang, De-Shuang Huang, Kang-Hyun Jo
Summary: This article introduces a facial-landmark detector based on a stacked hourglass network and residual networks, which improves accuracy by modifying hourglass modules and using 1 x 1 convolution layers in branch streams. The proposed network outperforms other state-of-the-art methods on 3-D face alignment datasets.
IEEE TRANSACTIONS ON INDUSTRIAL INFORMATICS
(2021)
Article
Engineering, Electrical & Electronic
Zhiqun Pan, Yongxiong Wang, Sunjie Zhang
Summary: This paper proposes a real-time framework for joint face detection and Facial Landmark Localization (FLL). It utilizes a fully convolutional network to predict the location of facial landmarks and face regions, and introduces a progressively pseudo labeling training method to eliminate the effect of inaccurate/noisy annotations. Two graph matching algorithms without learnable parameters are also proposed for completing the bottom-up face assembly.
SIGNAL PROCESSING-IMAGE COMMUNICATION
(2022)
Article
Computer Science, Information Systems
Le Quan Nguyen, Van Dung Pham, Yanfen Li, Hanxiang Wang, L. Minh Dang, Hyoung-Kyu Song, Hyeonjoon Moon
Summary: The conventional heatmap regression with deep networks has achieved success in landmark detection, but fails to fully exploit the overall structure of landmarks. This paper proposes a new landmark detection method that models landmarks as a graph structure, enabling the capture of the overall structure. Experimental results demonstrate the robustness of the proposed method.
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
Flavio H. de B. Zavan, Olga R. R. Bellon, Luciano Silva, Gerard G. Medioni
PATTERN RECOGNITION LETTERS
(2019)
Article
Food Science & Technology
Luisa Ozorio, Luciano P. Silva, Maura Prates, Carlos Bloch Jr, Cristina Y. Takeiti, Danillo Macedo Gomes, Jose Eduardo da Silva-Santos, Rosires Deliza, Ana Iraidy S. Brigida, Angela Furtado, Caroline Mellinger-Silva, Lourdes M. C. Cabral
FOOD RESEARCH INTERNATIONAL
(2019)
Article
Agriculture, Dairy & Animal Science
Thiago Felipe Braga, Thainara Christie Ferreira Silva, Mariana Groke Marques, Andressa Pereira de Souza, Daniela Albring, Luciano Paulino Silva, Alexandre Rodrigues Caetano, Margot Alves Nunes Dode, Mauricio Machaim Franco
REPRODUCTION IN DOMESTIC ANIMALS
(2019)
Article
Geochemistry & Geophysics
Rodrigo Minetto, Mauricio Pamplona Segundo, Sudeep Sarkar
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING
(2019)
Review
Computer Science, Artificial Intelligence
Gabriel Dahia, Leone Jesus, Mauricio Pamplona Segundo
WILEY INTERDISCIPLINARY REVIEWS-DATA MINING AND KNOWLEDGE DISCOVERY
(2020)
Article
Computer Science, Software Engineering
Beatriz Trinchao Andrade, Benjamin Resch, Hendrik P. A. Lensch, Olga Regina Pereira Bellon, Luciano Silva
COMPUTERS & GRAPHICS-UK
(2020)
Article
Computer Science, Artificial Intelligence
Marlon Marcon, Riccardo Spezialetti, Samuele Salti, Luciano Silva, Luigi Di Stefano
Summary: Correspondences between 3D keypoints generated by matching local descriptors play a crucial role in 3D computer vision and graphic applications. Learning effective representations of these keypoints requires supervised learning, which can be time-consuming and cumbersome. This paper proposes a method that learns equivariant 3D local descriptors, which outperform existing unsupervised methods and achieve competitive results in transfer learning scenarios.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Proceedings Paper
Computer Science, Artificial Intelligence
Andre Brasil Vieira Wyzykowski, Mauricio Pamplona Segundo, Rubisley de Paula Lemes
Summary: The study introduces a novel hybrid approach to create realistic, high-resolution fingerprints in response to legal restrictions on protecting biometric data privacy. The researchers managed to generate a synthetic database and included sweat pore annotations to encourage further research. The performance of real and synthetic databases was found to be similar, with human perception unable to differentiate between the two. The study suggests that the proposed approach is state-of-the-art in the field.
2020 25TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2021)
Article
Engineering, Electrical & Electronic
Mauricio Pamplona Segundo, Allan Pinto, Rodrigo Minetto, Ricardo da Silva Torres, Sudeep Sarkar
Summary: This work introduces a new method to measure economic activity through remote sensing, using signatures in satellite imagery left by disturbances in human behavior to devise image indicators for estimating the impact of major life events and supporting decision-makers. A case study on the COVID-19 outbreak was presented, analyzing the effects of lockdown measures around the 30 busiest airports in Europe through airplane detection and examining post-lockdown recovery. The solution won the RACE upscaling challenge, sponsored by the European Space Agency and the European Commission, and is now integrated into the RACE dashboard, utilizing satellite data and artificial intelligence for a safe reopening of essential activities.
IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND REMOTE SENSING
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Marlon Marcon, Olga Regina Pereira Bellon, Luciano Silva
Summary: This paper presents a novel pipeline for fine 6DoF pose estimation of objects in real-time, achieving high accuracy and fast processing rates. The modularity of the proposal allows for scheduled execution that unlocks real-time processing, even in multitask situations.
VISAPP: PROCEEDINGS OF THE 16TH INTERNATIONAL JOINT CONFERENCE ON COMPUTER VISION, IMAGING AND COMPUTER GRAPHICS THEORY AND APPLICATIONS - VOL. 5: VISAPP
(2021)
Article
Computer Science, Information Systems
Rodrigo Minetto, Mauricio Pamplona Segundo, Gilbert Rotich, Sudeep Sarkar
Summary: The COVID-19 outbreak led to global lockdowns and quarantines, disrupting human and economic activities worldwide. Recovery is expected to be difficult, as economic activities influence social behaviors, leaving traces in satellite images that can be automatically detected and classified. Satellite imagery provides a different type of visibility to support decision-making for analysts and policymakers.
IEEE TRANSACTIONS ON BIG DATA
(2021)
Proceedings Paper
Computer Science, Artificial Intelligence
Andre Roberto Ortoncelli, Luciano Silva, Olga Regina Perreira Bellon, Tiago Mota de Oliveira, Juliana Daga
2020 15TH IEEE INTERNATIONAL CONFERENCE ON AUTOMATIC FACE AND GESTURE RECOGNITION (FG 2020)
(2020)
Proceedings Paper
Computer Science, Artificial Intelligence
Luan P. E. Silva, Julio C. Batista, Olga R. P. Bellon, Luciano Silva
PROGRESS IN PATTERN RECOGNITION, IMAGE ANALYSIS, COMPUTER VISION, AND APPLICATIONS (CIARP 2019)
(2019)
Proceedings Paper
Computer Science, Software Engineering
Vitor Albiero, Olga R. P. Bellon, Luciano Silva
2018 25TH IEEE INTERNATIONAL CONFERENCE ON IMAGE PROCESSING (ICIP)
(2018)
Proceedings Paper
Computer Science, Artificial Intelligence
Raphael K. Czovny, Olga R. P. Bellon, Luciano Silva, Henrique S. G. Costa
2018 24TH INTERNATIONAL CONFERENCE ON PATTERN RECOGNITION (ICPR)
(2018)
