Boundary Attention: Learning curves, corners, junctions and grouping
This addresses the need for more accurate and rich boundary detection in computer vision, particularly for applications like image analysis, but it appears incremental as it builds on classical edge detection methods with learned components.
The paper tackles the problem of extracting detailed boundary structures like curves, corners, and junctions from images by introducing a lightweight network that uses boundary attention to refine geometric representations. It achieves generalization from synthetic shapes to noisy low-light photographs, though no concrete performance numbers are provided.
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.