2021 IEEE International Conference on Image Processing
19-22 September 2021 • Anchorage, Alaska, USA
Imaging Without Borders
| Paper ID | MLR-APPL-MDSP.8 | ||
| Paper Title | CYCLIC DIFFEOMORPHIC TRANSFORMER NETS FOR CONTOUR ALIGNMENT | ||
| Authors | Ilya Kaufman, Ron Shapira Weber, Oren Freifeld, Ben-Gurion University, Israel | ||
| Session | MLR-APPL-MDSP: Machine learning for multidimensional signal processing | ||
| Location | Area F | ||
| Session Time: | Monday, 20 September, 13:30 - 15:00 | ||
| Presentation Time: | Monday, 20 September, 13:30 - 15:00 | ||
| Presentation | Poster | ||
| Topic | Applications of Machine Learning: Machine learning for multidimensional signal processing | ||
| IEEE Xplore Open Preview | Click here to view in IEEE Xplore | ||
| Abstract | Shape analysis is a key task in image processing. A common method for representing a 2D shape is via a polygon, where the latter is a discretized version of the contour outlining the shape. However, due to the problem of curve reparameterization (\ie, points ``sliding'' along the contour), even if several shapes are very similar, their representations might be misleading far from each other. This misalignment problem confounds shape analysis. As a remedy, we propose a deep-learning framework, based on the recently-proposed diffeomorphic transformers nets. The proposed method handles either a single class (in an unsupervised manner) or multiple classes (in a semi-supervised manner), and is amenable to the warp-around effect exhibited in closed contours. Moreover, unlike typical alignment methods unrelated to learning, the proposed method aligns not only the original (``training'') shapes but also generalizes to test shapes (even if no class labels are given during the test). Our code is publicly available athttps://github.com/BGU-CS-VIL/CDTNCA | ||