10 selected Publications (2018-2023)

Articles de revues avec comité de lecture

1. A Generalized Asymmetric Dual-front Model for Active Contours and Image Segmentation.
   Da Chen, Jack Spencer, Jean-Marie Mirebeau, Ke Chen, Ming-Lei Shu and Laurent D. Cohen.
     in     IEEE Trans. on Image Processing, Volume 30, May 2021, Pages 5056-5071. 10.1109/TIP.2021.3078102t;. ;hal-03425885t;
2. Trajectory Grouping with Curvature Regularization for Tubular Structure Tracking. Li Liu, Da Chen, Ming-Lei Shu, Baosheng Li, Huazhong Shu, Michel Paques and Laurent D. Cohen.
    in     IEEE Trans. on Image Processing, Volume 31, 2021, pages 405 ? 418.  10.1109/TIP.2021.3131940t;. ;hal-02996874t;
3. Curvilinear Structure Tracking Based on Dynamic Curvature-penalized Geodesics. Li Liu, Mingzhu Wang, Shuwang Zhou, Minglei Shu, Laurent D. Cohen and Da Chen.
    in     Pattern Recognition, Volume 134, August  2023
   Pages 766-783.
    https://doi.org/10.1016/j.patcog.2022.109079
4. Computing geodesic paths encoding a curvature prior for curvilinear structure tracking. Da Chen, Jean-Marie Mirebeau, Minglei Shu and Laurent D. Cohen.
    in     PNAS, Proceedings of the National Academy of Sciences of the United States of America, Volume 120 (33), August 7, 2023. 10.1073/pnas.22188691
5. Geodesic Models With Convexity Shape Prior.
   Da Chen, Jean-Marie Mirebeau, Minglei Shu, Xuecheng  Tai  and Laurent D. Cohen.
     in     IEEE Transactions on Pattern Analysis and Machine Intelligence (PAMI), volume 45, no. 07, pp. 8433-8452, 2023.  https://doi.org/10.1007/s11263-023-01881-z
6. A Region-Based Randers Geodesic Approach for Image Segmentation.
   Da Chen, Jean-Marie Mirebeau, Huazhong Shu and Laurent D. Cohen.
     in     International Journal of Computer Vision (IJCV),  2023
    doi: 10.1109/TPAMI.2022.3225192  

Proceedings de colloques avec comité de lecture

1. Shape Morphing as a Minimal Path in the Graph of Cubified Shapes. Raphael Groscot and  Laurent D. Cohen.
    In   18th International Joint Conference on Computer Vision, Imaging
   and Computer Graphics Theory and Applications (VISIGRAPP23), February 19-21, 2023, Lisbon, Portugal. DOI: 10.5220/0011680200003417  Best Paper Award. 
2. Fast Marching Energy CNN. Learning to define a metric and a center such that the segmented region is the unit ball according to this metric and center.
   Theo Bertrand, Nicolas Makaroff and Laurent D. Cohen.
     In     Proc. 9th International Conference on Scale Space and Variational Methods in Computer Vision (SSVM 2023), Sardinia - Italy, May 21-25 2023.
3. A Model Is Worth Tens of Thousands of Examples.  We show with simple toy examples that in order for CNN methods to perform better than a model, tens of thousands of examples are needed for training.       Thomas Dages, Laurent D. Cohen and Alfred M. Bruckstein. In Proc. 9th International Conference on Scale Space and Variational Methods in Computer Vision (SSVM 2023)}}, Sardinia - Italy, May 21-25 2023.

Chapitres d'ouvrages

1. From Active Contours to Minimal Geodesic Paths: New Solutions to Active Contours Problems by Eikonal Equations. We propose different approaches based on Finsler metrics and mathematical tricks in order to minimize through geodesic methods the classical energy terms in active contour models. This takes advantage of geodesic methods which can find a global minimizer with fast method. Finding a shortest path according to relevant metrics and spaces allows to reproduce active contours terms for curvature penalization, alignment with a given vector field and region homogeneity term.              Da Chen and  Laurent D. Cohen, in Handbook of Numerical Analysis, volume 20-- Processing, Analyzing and Learning of Images, Shapes, and Forms, pages 1--44, Elsevier, 2019.