# Boundary Attention: Learning curves, corners, junctions and grouping

arXiv: https://arxiv.org/abs/2401.00935 · HTML: https://mlsanity.com/paper/2401.00935/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 7 of its 26 references that are in the map.
- First posted on arXiv 1 Jan 2024.

## Builds on (most cited first, 7 of 7)

- [MLP-Mixer: An all-MLP Architecture for Vision](https://mlsanity.com/paper/2105.01601/) (2021)
- [Holistically-Nested Edge Detection](https://mlsanity.com/paper/1504.06375/) (2015)
- [Pixel Difference Networks for Efficient Edge Detection](https://mlsanity.com/paper/2108.07009/) (2021)
- [A Physics-based Noise Formation Model for Extreme Low-light Raw Denoising](https://mlsanity.com/paper/2003.12751/) (2020)
- [Fast Edge Detection Using Structured Forests](https://mlsanity.com/paper/1406.5549/) (2014)
- [The Treasure Beneath Multiple Annotations: An Uncertainty-aware Edge Detector](https://mlsanity.com/paper/2303.11828/) (2023)
- [Field of Junctions: Extracting Boundary Structure at Low SNR](https://mlsanity.com/paper/2011.13866/) (2020)

## Built on by (most cited first, 1 of 1)

- [Non-Learning Low-Light Stereo Vision](https://mlsanity.com/paper/2606.00379/) (2026)

## Abstract (arXiv)

We present a lightweight network that infers grouping and boundaries, including curves, corners and junctions. It operates in a bottom-up fashion, analogous to classical methods for sub-pixel edge localization and edge-linking, but with a higher-dimensional representation of local boundary structure, and notions of local scale and spatial consistency that are learned instead of designed. Our network uses a mechanism that we call boundary attention: a geometry-aware local attention operation that, when applied densely and repeatedly, progressively refines a pixel-resolution field of variables that specify the boundary structure in every overlapping patch within an image. Unlike many edge detectors that produce rasterized binary edge maps, our model provides a rich, unrasterized representation of the geometric structure in every local region. We find that its intentional geometric bias allows it to be trained on simple synthetic shapes and then generalize to extracting boundaries from noisy low-light photographs.