|
Ph. D., INRIA Sophia Antipolis
Keywords: Curvilinear structure modeling, Stochastic process, Structure inference, Machine learning Contact: |
 |
Abstract:
My work at INRIA deals with the development of curvilinear structure reconstruction models based on stochastic modeling and machine learning. I assume that the curvilinear structure can be decomposed into a set of line segments. This assumption enables us to reconstruct arbitrary shapes of curvilinear structure for different types of datasets, e.g., retinal images, DNA filaments, facial wrinkles, road networks. I look for an optimal set of line segments which correspond to the latent curvilinear structure. For the stochastic model, I propose a probability density of the line segments for a given image data which constrains geometric interaction of line segments. Also, I employ a machine learning technique to infer structured information and to predict the correspondence of the given line segment and the latent curvilinear structures.
Short Bio:
Seong-Gyun Jeong received the B.E. degree in Electrical Engineering and M.E. degree in Electrical & Computer engineering from the Korea University, Seoul, Korea, in 2010 and 2012, respectively. Currently he is working as a Ph.D. student with AYIN team, INRIA. In particular, his research concerns stochastic image modeling, structure inference, machine learning, and graph theory.
Last publications:
2016
Preprints, Working Papers, …
- ref_biblio
- Seong-Gyun Jeong, Yuliya Tarabalka, Nicolas Nisse, Josiane Zerubia. Progressive Tree-like Curvilinear Structure Reconstruction with Structured Ranking Learning and Graph Algorithm. 2016. ⟨hal-01414864⟩
- Accès au texte intégral et bibtex
-
2015
Conference papers
- ref_biblio
- Seong-Gyun Jeong, Yuliya Tarabalka, Josiane Zerubia. Stochastic model for curvilinear structure reconstruction using morphological profiles. ICIP 2015 – IEEE International Conference on Image Processing, IEEE, Sep 2015, Quebec City, Canada. ⟨hal-01152932⟩
- Accès au texte intégral et bibtex
-
- ref_biblio
- Seong-Gyun Jeong, Yuliya Tarabalka, Josiane Zerubia. Marked Point Process Model for Curvilinear Structures Extraction. EMMCVPR 2015, Jan 2015, Hong Kong, Hong Kong SAR China. pp.436-449, ⟨10.1007/978-3-319-14612-6_32⟩. ⟨hal-01084939⟩
- Accès au bibtex
-
Reports
- ref_biblio
- Seong-Gyun Jeong, Yuliya Tarabalka, Nicolas Nisse, Josiane Zerubia. Inference of Curvilinear Structure based on Learning a Ranking Function and Graph Theory. [Research Report] RR-8789, Inria Sophia Antipolis. 2015. ⟨hal-01214932⟩
- Accès au texte intégral et bibtex
-
Theses
- ref_biblio
- Seong-Gyun Jeong. Curvilinear structure modeling and its applications in computer vision. Other [cs.OH]. Université Nice Sophia Antipolis, 2015. English. ⟨NNT : 2015NICE4086⟩. ⟨tel-01243028v2⟩
- Accès au texte intégral et bibtex
-
2014
Conference papers
- ref_biblio
- Seong-Gyun Jeong, Yuliya Tarabalka, Josiane Zerubia. Marked point process model for facial wrinkle detection. IEEE ICIP – International Conference on Image Processing, Oct 2014, Paris, France. pp.1391-1394. ⟨hal-01066231⟩
- Accès au texte intégral et bibtex
-
