Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

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.

Mengyang Pu, Yaping Huang, Yuming Liu, Qingji Guan, Haibin Ling• 2022

Related benchmarks

TaskDatasetResultRank
Boundary DetectionBSDS 500 (test)
ODS84.8
185
Boundary DetectionBSDS500
ODS F-score0.84
37
Edge DetectionBSDS v1 (test)
ODS82.4
32
Boundary DetectionMulticue Boundary
ODS86.1
26
Edge DetectionNYUD Standard Evaluation - SEval v2 (val)
ODS77.4
17
Edge DetectionNYUD v2
ODS0.774
16
Edge DetectionBIPED
ODS Score89.3
14
Edge DetectionMulti-Cue
ODS89.4
13
Edge DetectionNYUD Crispness-emphasized evaluation - CEval v2 (val)
ODS Score43
13
Boundary DetectionMulti-Cue (test)
ODS0.861
12
Showing 10 of 12 rows

Other info

Follow for update