Back to Search
Start Over
Differential invariants for SE(2)-equivariant networks
- Source :
- 29th IEEE International Conference on Image Processing (IEEE ICIP), Oct 2022, Bordeaux, France
- Publication Year :
- 2022
-
Abstract
- Symmetry is present in many tasks in computer vision, where the same class of objects can appear transformed, e.g. rotated due to different camera orientations, or scaled due to perspective. The knowledge of such symmetries in data coupled with equivariance of neural networks can improve their generalization to new samples. Differential invariants are equivariant operators computed from the partial derivatives of a function. In this paper we use differential invariants to define equivariant operators that form the layers of an equivariant neural network. Specifically, we derive invariants of the Special Euclidean Group SE(2), composed of rotations and translations, and apply them to construct a SE(2)-equivariant network, called SE(2) Differential Invariants Network (SE2DINNet). The network is subsequently tested in classification tasks which require a degree of equivariance or invariance to rotations. The results compare positively with the state-of-the-art, even though the proposed SE2DINNet has far less parameters than the compared models.
- Subjects :
- Electrical Engineering and Systems Science - Image and Video Processing
Subjects
Details
- Database :
- arXiv
- Journal :
- 29th IEEE International Conference on Image Processing (IEEE ICIP), Oct 2022, Bordeaux, France
- Publication Type :
- Report
- Accession number :
- edsarx.2206.13279
- Document Type :
- Working Paper