Learning to Discover Reflection Symmetry via Polar Matching Convolution

TL;DR

This paper introduces a novel polar matching convolution technique that improves reflection symmetry detection in images by leveraging specialized pooling, self-similarity encoding, and a systematic kernel design, outperforming existing methods.

Contribution

The paper proposes a new convolutional approach tailored for symmetry detection, addressing limitations of standard convolutions and introducing a new dataset with self-supervised training strategies.

Findings

Outperforms state-of-the-art symmetry detection methods.

Effective in real-world images with variations and ambiguities.

Robustness demonstrated through extensive experiments.

Abstract

The task of reflection symmetry detection remains challenging due to significant variations and ambiguities of symmetry patterns in the wild. Furthermore, since the local regions are required to match in reflection for detecting a symmetry pattern, it is hard for standard convolutional networks, which are not equivariant to rotation and reflection, to learn the task. To address the issue, we introduce a new convolutional technique, dubbed the polar matching convolution, which leverages a polar feature pooling, a self-similarity encoding, and a systematic kernel design for axes of different angles. The proposed high-dimensional kernel convolution network effectively learns to discover symmetry patterns from real-world images, overcoming the limitations of standard convolution. In addition, we present a new dataset and introduce a self-supervised learning strategy by augmenting the dataset with synthesizing images. Experiments demonstrate that our method outperforms state-of-the-art methods in terms of accuracy and robustness.