Boundary Attention: Learning curves, corners, junctions and grouping
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866910604956008448 |
|---|---|
| author | Polansky, Mia Gaia Herrmann, Charles Hur, Junhwa Sun, Deqing Verbin, Dor Zickler, Todd |
| author_facet | Polansky, Mia Gaia Herrmann, Charles Hur, Junhwa Sun, Deqing Verbin, Dor Zickler, Todd |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2401_00935 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Boundary Attention: Learning curves, corners, junctions and grouping Polansky, Mia Gaia Herrmann, Charles Hur, Junhwa Sun, Deqing Verbin, Dor Zickler, Todd Computer Vision and Pattern Recognition 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. |
| title | Boundary Attention: Learning curves, corners, junctions and grouping |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2401.00935 |