Thanh Phuong Nguyen

Organization and administrative tasks

• 2024-now : Member of INPS pole of university of Toulon
• 2024-now: Member of Commission of communication, LIS, UMR 7020
• 2023-now: Co-Head of SIIM team, LIS, UMR 7020
• 2023: Delegate of LIS laboratory at INPS pole of University of Toulon
• 2023-now: Member of Commission of research support, LIS, UMR 7020
• 2023: I organized the conference ORASIS'23
• 2018-now: Member of LIS laboratory committee

Projects

I am the principal investigator of the following projects.

• 2023: Project ORASIS'23: Subvention for organization of ORASIS'23 (9 k€)
• 2022: Project CARTT (2,2 k€)
• 2022-2024: Project ROV-Chasseur: ANR Astrid 2021 on Robotic (300 k€)
• 2021: Project BQR LIS (3 k€)

I have been involved in the following projects.

• 2021-2022: RAPID project (Manta)
• 2018-2020: RAPID project (Sankit)
• 2013-2015: Digeteo project
• 2012-2013: European ITEA2 project (SPY)
• 2011-2012: Confidential industrial project
• 2007-2010: ANR white project (GeoDib)

Supervision

PhD students

Defended

• Thanh Tuan Nguyen (2017-2021 with Frederic Bouchara, University of Toulon)

In progress

• Cong Tu TRAN, (from 2023)
• Thuy Han Uyen THAI, (from 2023, Aix-Marseille University)
• Thi Quynh Khanh DINH, (from 2023 with Nadège Thirion-Moreau, University of Toulon, and Jean-Jacques Simon, Aix-Marseille University)
• Van Tien PHAM (from 2022 with Yassine Zniyed, University of Toulon, France)
• Yassimine SOUSSI (from 2021 with Mohamed Ben Khelifa, Ali Wali, Nizar Rokbani, University of Sousa, Tunisia)
• Ngoc Phuong Truong (from 2022 with Dinh Duy Le, VNU University, Viet Nam)
• Maurgène Ghezal (from 2021 with Jean Jaques Simon, Aix-Marseille University, France)

Postdoc

• Hong Hiep Nguyen
• Thanh Tuan Nguyen

PhD committees

• Hai Hong Phan, 2019, Université de Cergy Pontoise, France (examinator)
• Phuoc Hung Truong, 2021, Sejong University, Korea (examinator)

Mid-term PhD committees

• 2023: Sandra Martin, Aix-Marseille University, France
• 2021, 2022 : Ghader Bilal, University of Toulon, France
• 2022: Souhir Khessiba, University of Paris Saclay, France

Research activities

My research concentrates on different domains of computer vision but it can be resumed by these following topics

• Discrete geometry and Image analysis
• Shape analysis
• Texture analysis
• Human action recognition
• Face analysis
• Object detection

Discrete geometry and Image analysis

I have worked on this topic specially during my PhD thesis. I was interested in the study of discrete curves and its applications in image analysis. Three main approaches have been studied.

• Geometric estimators of digital curves : I have proposed an amelioration of curvature estimation based on circumcircle. This method is based on the notion of blurred segment of width v and on the decomposition of a curve into the sequence of maximal blurred segment of width v. Afterwards, I have applied this idea in 3D to estimate the discrete curvature and torsion at each point of a 3D curve. Main results:
• On the local properties of digital curves, Thanh Phuong Nguyen and Isabelle Debled Rennesson, International Journal of Shape Modeling, volume 14, issue 2, 2008, pp 105-125
• Curvature and torsion estimator for 3D curves, Thanh Phuong Nguyen and Isabelle Debled Rennesson, International Symposium on Visual Computing (ISVC'08)
• Curvature estimation in noisy curves, Thanh Phuong Nguyen and Isabelle Debled Rennesson, International Conference on Computer Analysis of Images and Patterns (CAIP'07)
• Dominant point detection: I have developed a rapid et reliable method to detect dominant points of a 2D curve. A dominant point is a point whose the curvature value is locally maximum. The dominant points play an important role in pattern recognition. Our method uses a parameter: the width of maximal blurred segments. Based on this novel method of dominant point detection, I proposed free-parameter methods for polygonal representation. They are based on a multi-width approach. Main results:
• A discrete geometry approach for dominant point detection, Thanh Phuong Nguyen and Isabelle Debled Rennesson, Pattern Recognition, 2011
• A multiscale approach to decompose a digital curve into meaningful parts, Thanh Phuong Nguyen and Isabelle Debled Rennesson, International Conference on Pattern Recognition (ICPR'10)
Incorporating two first order moments into LBP-based operator for texture categorization, Thanh Phuong Nguyen and Antoine Manzanera, ACCV Workshops, 2014
• Fast and robust dominant point detection on digital curves, Thanh Phuong Nguyen and Isabelle Debled Rennesson, International Conference on Image Processing (ICIP'09)
• Free-parameter method for polygonal representation of noisy curves, Thanh Phuong Nguyen and Isabelle Debled Rennesson, International Workshop on Combinatorial Image Analysis (IWCIA'09)
• Discerete arcs and circles: Finally, I'm interested in discrete arcs and circles. A linear method has been proposed for the recognition of arcs and circles. I then develop a new method for segmentation of noisy curves into arcs based on a method of noise detection. I also proposed a linear method to measure the circularity of closed curves. In addition, I have proposed a robust method to decompose a curve into arcs and line segments. Main results:
• Ellipse detection through decomposition of circular arcs and line segments, Thanh Phuong Nguyen and Bertrand Kerautret, International Conference on Image Analysis and Processing (ICIAP'11)
• Arc segmentation in linear time, Thanh Phuong Nguyen and Isabelle Debled-Rennesson, International Conference on Computer Analysis of Images and Patterns (CAIP'11)
• Decomposition of a curve into arcs and line segments based on dominant point detection, Thanh Phuong Nguyen and Isabelle Debled-Rennesson, Scandinavian Conference on Image Analysis (SCIA'11)
• Circular arc reconstruction of digital contours with chosen Hausdorff error, Bertrand Kerautret, Jaques Oliver Lachaud and Thanh Phuong Nguyen, Discrete Geometry for Computer Imagery (DGCI'11)
• Circularity measuring in linear time, Thanh Phuong Nguyen and Isabelle Debled Rennesson, International Conference on Pattern Recognition (ICPR'10)
• Unsupervised, Fast and Precise Recognition of Digital Arcs in Noisy Images, Thanh Phuong Nguyen, Bertrand Kerautret, Isabelle Debled Rennesson and Jaques Oliver Lachaud, International Conference on Computer Vision and Graphics (ICCVG'10)

Shape analysis

Beside discrete geometry, I'm also interested in other approaches for shape analysis. Recently, I developped several tools based on Radon transform to fulfill this purpose. Main result:

• Projection-based Polygonality Measurement, Thanh Phuong Nguyen and Van-Thai Hoang, TIP'15
• Shape measurement using LIP-signature, Thanh Phuong Nguyen, Xuan Son Nguyen, CVIU'18
• Line and circle detection using dense one-to-one Hough transforms on greyscale images, Antoine Manzanera, Thanh Phuong Nguyen, Xiaolei Xu, EURASIP Journal on Image and Video Processing, 2016
• A projection based method for shape measurement, Thanh Phuong Nguyen, Xuan Son Nguyen, Mohamed Anouar Borgi, M. K. Nguyen, Journal of Mathematical Imaging and Vision, 2020
• Reflection symmetry detection of shapes based on shape-signatures, Thanh Phuong Nguyen, Phuoc Hung Truong, Y. Kim, Thanh Tuan Nguyen, Pattern Recognition, 2022 (in minor revision)

Texture analysis

Texture analysis has been an active research topic after my PhD thesis. I'm mainly interested in texture classification. Several aspects have been studied including preprocessing, texture descriptors, LBP-based representation and so on. Main results:

• Incorporating two first order moments into LBP-based operator for texture categorization, Thanh Phuong Nguyen and Antoine Manzanera, ACCV Workshops, 2014
• Improving Texture Categorization with Biologically Inspired Filtering, Ngoc-Son Vu, Thanh Phuong Nguyen, Christophe Garcia, Image and Vision Computing, 2014
• Impact of rotation invariant attributes from Local Binary Patterns on texture classification, Thanh Phuong Nguyen, Antoine Manzanera and Walter G. Kropatsch, ECCV Workshops, 2014
• Topological Attribute Patterns for Texture Recognition, Thanh Phuong Nguyen, Antoine Manzanera, Walter G. Kropatsch and Xuan-Son Nguyen, Pattern Recognition Letters, volume 80, 2016
• Statistical Binary Patterns for Rotation Invariant Texture Classification, Thanh Phuong Nguyen, Ngoc-Son Vu and Antoine Manzanera, Neurocomputing, volume 173, 2016

Dynamic texture recognition

• Dynamic Texture Representation based on Oriented Magnitudes of Gaussian Gradients, Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frederic Bouchara, Journal of Visual Communication and Image Representation, 2021
• A Comprehensive Taxanomy of Dynamic Texture Representation, Thanh Tuan Nguyen, Thanh Phuong Nguyen, ACM Computing Surveys, 2021
• Dynamic Texture Description Using Adapted Bipolar-Invariant and Blurred Features, Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frederic Bouchara, Multidimensional Systems and Signal Processing, 2022
• A Novel Filtering Kernel based on Difference of Derivative Gaussians with Applications to Dynamic Texture Representation, Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frederic Bouchara, Signal Processing: Image Communication, 2021
• Prominent Local Representation for Dynamic Textures Based on High-order Gaussian-gradients, Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frederic Bouchara, IEEE Transactions on Multimedia, 2021
• Rubic Gaussian-based patterns for dynamic texture classification, Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frederic Bouchara, Pattern Recognition Letters, 2020
• Directional Dense-Trajectory-based Patterns for Dynamic Texture Recognition, Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frederic Bouchara, IET Computer Vision, 2020
• Momental directional patterns for dynamic texture recognition, Thanh Tuan Nguyen, Thanh Phuong Nguyen, Frederic Bouchara, Xuan Son Nguyen, Computer Vision and Image Understanding, 2020
• Completed Statistical Adaptive Patterns on Three Orthogonal Planes for Recognition of Dynamic Textures and Scenes, Thanh Tuan Nguyen, Thanh Phuong Nguyen, and Frédéric Bouchara, Journal of Electronic Imaging, volume 27, number 5, 2018

Human action recognition

In the framework of project SPY, I was insterested much more in human activity analysis. Our purpose is to design an efficient descriptor for human action recognition that can be intergrated in an embeded system having a limited computational capability. In addition, a such system can work in near real-time condition. Main results

• Action Recognition using Bag of Features Extracted from a Beam of Trajectories, Thanh Phuong Nguyen and Antoine Manzanera, International Conference on Image Processing (ICIP'13)
• Motion Trend Patterns for Action Modelling and Recognition, Thanh Phuong Nguyen, Antoine Manzanera and Matthieu Garrigues, Computer Analysis of Images and Patterns (CAIP'13)
• Revisiting LBP-based Texture Models for Human Action Recognition, Thanh Phuong Nguyen, Antoine Manzanera, Ngoc-Son Vu and Matthieu Garrigues, Iberoamerican Congress on Pattern Recognition (CIARP'13)
• Spatial Motion Patterns: Action Models from semi-dense Trajectories, Thanh Phuong Nguyen, Antoine Manzanera, Ngoc-Son Vu and Matthieu Garrigues, IJPRAI, 2014
• Local Derivative Pattern for Action Recognition in Depth Images , Xuan-Son Nguyen, Thanh Phuong Nguyen, François Charpillet, Ngoc Son Vu, Multimedia Tools and Applications, 2018
• Action Recognition in Depth Videos Using Hierarchical Gaussian Descriptor, Xuan Son Nguyen, Abdel-Illah Mouaddib, Thanh Phuong Nguyen, Laurent Jeanpierre, Multimedia Tools and Applications, 2018
• Hierarchical Gaussian Descriptor Based on Local Pooling for Action Recognition, Xuan Son Nguyen, Abdel-Illah Mouaddib, and Thanh Phuong Nguyen, Machine Vision and Applications, 2019

Face analysis

I have considered the problem of face recognition as an application of texture descriptor. Main result:

• Weighted Statistical Binary Patterns for Facial Feature Representation, Phuoc Hung Truong, Thanh Phuong Nguyen, Yong-Guk Kim, Applied Intelligence, Springer, 2021
• Statistical binary patterns and post-competitive representation for pattern recognition, Mohamed Anouar Borgi, Thanh Phuong Nguyen, Demetrio Labate, Chokri Ben Amar, International Journal of Machine Learning and Cybernetics, 2018
• Face Recognition using Multi-modal Binary Patterns, Thanh Phuong Nguyen, Ngoc-Son Vu and Alice Caplier, International Conference on Pattern Recognition (ICPR'12)