Model card
DBNet++ (ResNet-18) (736).
UnknownunknownUnknown paramsUnknown
Imported from Papers With Code
§ 02 · Benchmarks
Every benchmark DBNet++ (ResNet-18) (736) has a recorded score for.
| # | Benchmark | Area · Task | Metric | Value | Rank | Date | Source |
|---|---|---|---|---|---|---|---|
| 01 | ICDAR 2015 | Computer Vision · Scene Text Detection | fps | 44.0% | #3 | 2022-02-21 | source ↗ |
| 02 | msra-td500 | Computer Vision · Scene Text Detection | fps | 55.0% | #6 | 2022-02-21 | source ↗ |
| 03 | msra-td500 | Computer Vision · Scene Text Detection | recall | 82.5% | #10 | 2022-02-21 | source ↗ |
| 04 | msra-td500 | Computer Vision · Scene Text Detection | f-measure | 85.1% | #11 | 2022-02-21 | source ↗ |
| 05 | ICDAR 2015 | Computer Vision · Scene Text Detection | precision | 90.1% | #16 | 2022-02-21 | source ↗ |
| 06 | msra-td500 | Computer Vision · Scene Text Detection | precision | 87.9% | #16 | 2022-02-21 | source ↗ |
| 07 | ICDAR 2015 | Computer Vision · Scene Text Detection | f-measure | 83.1% | #32 | 2022-02-21 | source ↗ |
| 08 | ICDAR 2015 | Computer Vision · Scene Text Detection | recall | 77.2% | #36 | 2022-02-21 | 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.
§ 03 · Strengths by area
Where DBNet++ (ResNet-18) (736) actually performs.
§ 04 · Papers
1 paper with results for DBNet++ (ResNet-18) (736).
- 2022-02-21· Computer Vision· 8 results
Real-Time Scene Text Detection with Differentiable Binarization and Adaptive Scale Fusion
§ 05 · Related models
Other Unknown models scored on Codesota.
CRAFT
Unknown params · 21 results
GPT-2-Large (fine-tuning)
Unknown params · 20 results
HTLM (fine-tuning)
Unknown params · 20 results
TextFuseNet (ResNeXt-101)
Unknown params · 16 results
MixNet
Unknown params · 13 results
PAN
Unknown params · 12 results
SPCNET
Unknown params · 12 results
TESTR
Unknown params · 12 results
§ 06 · Sources & freshness
Where these numbers come from.
papers-with-code
8
results
8 of 8 rows marked verified.