RotaTouille: Rotation Equivariant Deep Learning for Contours
Abstract
RotaTouille is a deep learning framework that achieves rotation and cyclic shift equivariance for contour data using complex-valued circular convolution, enabling effective performance in shape classification, reconstruction, and contour regression.
Contours or closed planar curves are common in many domains. For example, they appear as object boundaries in computer vision, isolines in meteorology, and the orbits of rotating machinery. In many cases when learning from contour data, planar rotations of the input will result in correspondingly rotated outputs. It is therefore desirable that deep learning models be rotationally equivariant. In addition, contours are typically represented as an ordered sequence of edge points, where the choice of starting point is arbitrary. It is therefore also desirable for deep learning methods to be equivariant under cyclic shifts. We present RotaTouille, a deep learning framework for learning from contour data that achieves both rotation and cyclic shift equivariance through complex-valued circular convolution. We further introduce and characterize equivariant non-linearities, coarsening layers, and global pooling layers to obtain invariant representations for downstream tasks. Finally, we demonstrate the effectiveness of RotaTouille through experiments in shape classification, reconstruction, and contour regression.
Community
A deep learning framework for contour data with layers equivariant and invariant to planar rotations and cyclic shifts. Based on complex-valued neural networks, making it easy to implement in existing libraries such as, for example, PyTorch.
Figure: Original input contour and reconstruction for a baseline image-based CNN and our method. Notice how, by equivariance, the stabilized reconstruction stays constant, whereas for the baseline CNN, we observe some deformations happening.
Implementation available at GitHub: https://github.com/odinhg/rotation-equivariant-contour-learning
This is an automated message from the Librarian Bot. I found the following papers similar to this paper.
The following papers were recommended by the Semantic Scholar API
- Equivariant Volumetric Grasping (2025)
- Axis-level Symmetry Detection with Group-Equivariant Representation (2025)
- ShapeEmbed: a self-supervised learning framework for 2D contour quantification (2025)
- Representation Theory of $UT_3(\mathbb{F}_3)$ and its Applications to Equivariant Decomposition in Neural Architectures (2025)
- Measuring the Impact of Rotation Equivariance on Aerial Object Detection (2025)
- Rotation Equivariant Arbitrary-scale Image Super-Resolution. (2025)
- Local Scale Equivariance with Latent Deep Equilibrium Canonicalizer (2025)
Please give a thumbs up to this comment if you found it helpful!
If you want recommendations for any Paper on Hugging Face checkout this Space
You can directly ask Librarian Bot for paper recommendations by tagging it in a comment:
@librarian-bot
recommend
Models citing this paper 0
No model linking this paper
Datasets citing this paper 0
No dataset linking this paper
Spaces citing this paper 0
No Space linking this paper
Collections including this paper 0
No Collection including this paper