EDTER: Edge Detection with Transformer
About
Convolutional neural networks have made significant progresses in edge detection by progressively exploring the context and semantic features. However, local details are gradually suppressed with the enlarging of receptive fields. Recently, vision transformer has shown excellent capability in capturing long-range dependencies. Inspired by this, we propose a novel transformer-based edge detector, \emph{Edge Detection TransformER (EDTER)}, to extract clear and crisp object boundaries and meaningful edges by exploiting the full image context information and detailed local cues simultaneously. EDTER works in two stages. In Stage I, a global transformer encoder is used to capture long-range global context on coarse-grained image patches. Then in Stage II, a local transformer encoder works on fine-grained patches to excavate the short-range local cues. Each transformer encoder is followed by an elaborately designed Bi-directional Multi-Level Aggregation decoder to achieve high-resolution features. Finally, the global context and local cues are combined by a Feature Fusion Module and fed into a decision head for edge prediction. Extensive experiments on BSDS500, NYUDv2, and Multicue demonstrate the superiority of EDTER in comparison with state-of-the-arts.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Boundary Detection | BSDS 500 (test) | ODS84.8 | 185 | |
| Boundary Detection | BSDS500 | ODS F-score0.84 | 37 | |
| Edge Detection | BSDS v1 (test) | ODS82.4 | 32 | |
| Boundary Detection | Multicue Boundary | ODS86.1 | 26 | |
| Edge Detection | NYUD Standard Evaluation - SEval v2 (val) | ODS77.4 | 17 | |
| Edge Detection | NYUD v2 | ODS0.774 | 16 | |
| Edge Detection | BIPED | ODS Score89.3 | 14 | |
| Edge Detection | Multi-Cue | ODS89.4 | 13 | |
| Edge Detection | NYUD Crispness-emphasized evaluation - CEval v2 (val) | ODS Score43 | 13 | |
| Boundary Detection | Multi-Cue (test) | ODS0.861 | 12 |