Model card
DPNet (ResNet-50, 736px).
Fang et al.open-sourceUnknown paramsResNet-50 + Channel Enhanced Self-Attention Module (CESAM) + Spatial Enhanced Self-Attention Module (SESAM)
Dual Perspective CNN-Transformer for scene text detection. Integrates CESAM and SESAM into ResNet backbone. 736×736 input. PLOS ONE 2024. DOI:10.1371/journal.pone.0309286.
§ 01 · Benchmarks
Every benchmark DPNet (ResNet-50, 736px) has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | msra-td500 | Computer Vision · Scene Text Detection | f-measure | 86.7% | #9 | 2024-10-15 | source ↗ |
| 02 | msra-td500 | Computer Vision · Scene Text Detection | precision | 91.4% | #10 | 2024-10-15 | source ↗ |
| 03 | msra-td500 | Computer Vision · Scene Text Detection | recall | 82.5% | #11 | 2024-10-15 | source ↗ |
Rank column shows this model’s position vs all other models scored on the same benchmark + metric (competitors after the slash). #1 in red means current SOTA. Sorted by rank, then newest result.
§ 02 · Strengths by area
Where DPNet (ResNet-50, 736px) actually performs.
§ 03 · Papers
1 paper with results for DPNet (ResNet-50, 736px).
- 2024-10-15· Computer Vision· 3 results
DPNet: Scene Text Detection Based on Dual Perspective CNN-Transformer
§ 04 · Related models
Other Fang et al. models scored on Codesota.
§ 05 · Sources & freshness
Where these numbers come from.
arxiv
3
results
3 of 3 rows marked verified.