# Classical Versus Deep Mirror-Symmetry Scoring: A Benchmark of Thirteen Methods

arXiv: https://arxiv.org/abs/2607.08379 · HTML: https://mlsanity.com/paper/2607.08379/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 8 of its 54 references that are in the map.
- First posted on arXiv 9 Jul 2026.

## Builds on (most cited first, 8 of 8)

- [MambaOut: Do We Really Need Mamba for Vision?](https://mlsanity.com/paper/2405.07992/) (2024)
- [Reflection and Rotation Symmetry Detection via Equivariant Learning](https://mlsanity.com/paper/2203.16787/) (2022)
- [A convolutional approach to reflection symmetry](https://mlsanity.com/paper/1609.05257/) (2016)
- [SRN: Side-output Residual Network for Object Symmetry Detection in the Wild](https://mlsanity.com/paper/1703.02243/) (2017)
- [Wavelet-based Reflection Symmetry Detection via Textural and Color Histograms](https://mlsanity.com/paper/1707.02931/) (2017)
- [Axis-level Symmetry Detection with Group-Equivariant Representation](https://mlsanity.com/paper/2508.10740/) (2025)
- [CLIPSym: Delving into Symmetry Detection with CLIP](https://mlsanity.com/paper/2508.14197/) (2025)
- [A toolbox for calculating objective image properties in aesthetics research](https://mlsanity.com/paper/2408.10616/) (2024)

## Abstract (arXiv)

Quantifying how mirror-symmetric an image is about a given axis (symmetry scoring) underpins applications from visual aesthetics to medical imaging, yet proposed scoring methods have never been compared on a common, statistically grounded protocol. We benchmark 13 scoring methods (9 collected from the literature; 4 introduced here) spanning from classical features to frozen deep features, across four single-axis and five multi-axis datasets under a reflection-exact protocol with a chance-anchored, significance-tested discrimination skill. Deep backbones perform best on single-axis and harder multi-axis protocols. However, a classical histogram-of-oriented-gradients (HOG) descriptor trails the best frozen-network readout by a small (but significant) margin, is not statistically separable from the runner-up (a CNN-filter measure), and runs ~300× faster on CPU. Our results show that discrimination concentrates in mid-scale oriented features, where deep backbones peak at a low or mid stage, and HOG peaks at a mid cell size. Among existing methods, frozen deep features thus offer little over a tuned classical descriptor for measuring symmetry; whether task-trained deep scorers can do better remains open. We release the scorers and harness, in imgsym, an open toolkit for image symmetry detection and measurement.