Deep Snake for instance segmentation

Encadrants : Vannary Meas-Yedid
Disponible : NON
Spécialité : IMA
Nombre d'étudiants : 1
Description :

Image segmentation is a fundamental and challenging problem in image processing and has been widely studied and developed. Recently, image segmentation has been performed based on deep learning approach in combination with active contour models and shows very interesting results.

Travail demandé : After a bibliographic study on image segmentation based on active contour models and deep learning the work consists in experimenting a Deep Snake model on biological image datasets with a special focus on touching objects. A comparison should be done with different existing methods of active contours, deep learning and deep snakes.
Liens complémentaires : - T. F. Chan and L. A. Vese, “Active contours without edges,” IEEE Transactions on Image Processing, vol. 10, no. 2, pp. 266–277, 2001.
- Diego Marcos, Devis Tuia, Benjamin Kellenberger, Lisa Zhang, Min Bai, Renjie Liao, and Raquel Urtasun. Learning deep structured active contours end-to-end. In CVPR, 2018
- T. H. N. Le, K. G. Quach, K. Luu, C. N. Duong, and M. Savvides, “Reformulating level sets as deep recurrent neural network approach to semantic segmentation,” IEEE Transactions on Image
Processing, vol. 27, no. 5, pp. 2393–2407, 2018.
- Sida Peng, Wen Jiang, Huaijin Pi, Xiuli Li, Hujun Bao, Xiaowei Zhou, Deep Snake for Real-Time Instance Segmentation, CVPR 2020