End-to-end scene text spotting on Total-Text
84.3Hmean (None)OmniParser V2
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| OmniParser V2FPS=8.22025.02 | 84.3 | 89.5 | — | — | — | |
| OmniParser V2 SPTpre-training=SPT, FPS=8.22025.02 | 84.2 | 89.3 | — | — | — | |
| OmniParser V2 PTpre-training=PT, FPS=8.22025.02 | 84.1 | 89.1 | — | — | — | |
| OmniParserFPS=2.62025.02 | 84 | 88.9 | — | — | — | |
| DeepSoloBackbone=ViTAEv2-S, External Data=Synth150K+MLT17+IC13+IC15+TextOCR, FPS (report)=10, FPS (A100)=102023.05 | 83.6 | 89.6 | — | — | — | |
| BridgeSpotterFPS=6.72025.02 | 83.3 | 88.3 | — | — | — | |
| DeepSoloBackbone=Res-50, External Data=Synth150K+MLT17+IC13+IC15+TextOCR, FPS (report)=17, FPS (A100)=172023.05 | 82.5 | 88.7 | — | — | — | |
| DeepSoloFPS=17.02025.02 | 82.5 | 88.7 | — | — | — | |
| UNITSFPS=3.12025.02 | 82.2 | 88 | — | — | — | |
| SRSTS v2 (ours)Scale=S: 6402022.07 | 82.05 | 88.05 | — | 12.86 | 12.86 | |
| DeepSolocode_source=open-source, FPS=17.02025.02 | 81.2 | 87.8 | — | — | — | |
| GLASSExternal Data=Synth800K, FPS (report)=32023.05 | 79.9 | 86.2 | — | — | — | |
| GLASSFPS=2.72025.02 | 79.9 | 86.2 | — | — | — | |
| SRSTS v2-F (ours)Scale=S: 6402022.07 | 79.75 | 87.05 | — | 20.22 | 20.22 | |
| DeepSoloBackbone=Res-50, External Data=Synth150K+MLT17+IC13+IC15, FPS (report)=17, FPS (A100)=172023.05 | 79.7 | 87 | — | — | — | |
| ABINet++Character-level annotations=false, Multi-scale=true2022.11 | 79.4 | 85.4 | — | — | — | |
| SRSTS (ours)Scale=S: 6402022.07 | 78.8 | 86.33 | — | 18.74 | 18.74 | |
| DeepSoloBackbone=Res-50, External Data=Synth150K, FPS (report)=17, FPS (A100)=172023.05 | 78.8 | 86.2 | — | — | — | |
| CraftYear=2020, Character-level annotations=true, Multi-scale=false2022.11 | 78.7 | — | 5.4 | — | — | |
| CRAFTSScale=L: 1920, Character-level annotations=true2022.07 | 78.7 | — | — | — | — | |
| CRAFTSExternal Data=Synth800K+IC13, Character-level Annotations=true2023.05 | 78.7 | — | — | — | — | |
| CRAFTSFPS=8.82025.02 | 78.7 | — | — | — | — | |
| TTS (poly)External Data=Synth800K+COCO-Text+IC13+IC15+SCUT2023.05 | 78.2 | 86.3 | — | — | — | |
| TTS2025.02 | 78.2 | 86.3 | — | — | — | |
| ABINet++Character-level annotations=false, Multi-scale=false2022.11 | 77.6 | 84.5 | 10.6 | — | — | |
| ABINet++External Data=Synth150K+MLT17, FPS (report)=10.62023.05 | 77.6 | 84.5 | — | — | — | |
| ABINet++FPS=10.62025.02 | 77.6 | 84.5 | — | — | — | |
| GLASS2022.07 | 76.6 | 83 | — | — | — | |
| TPSNetExternal Data=Synth150K+MLT17, Backbone=Res-50-DCN, FPS (report)=9.32023.05 | 76.1 | 82.3 | — | — | — | |
| TPSNet2025.02 | 76.1 | 82.3 | — | — | — | |
| TTSpoly2022.07 | 75.6 | 84.4 | — | — | — | |
| SwinTextspotterScale=S: 10002022.07 | 74.3 | 84.1 | — | — | — | |
| SwinTSBackbone=Swin-T, External Data=Synth150K+MLT17+IC13+IC15, FPS (report)=2.92023.05 | 74.3 | 84.1 | — | — | — | |
| SwinTextSpotterFPS=2.92025.02 | 74.3 | 84.1 | — | — | — | |
| SPTSExternal Data=Synth150K+MLT17+IC13+IC15, FPS (report)=0.62023.05 | 74.2 | 82.4 | — | — | — | |
| SPTSFPS=0.62025.02 | 74.2 | 82.4 | — | — | — | |
| TESTRExternal Data=Synth150K+MLT17, FPS (report)=5.3, FPS (A100)=12.12023.05 | 73.3 | 83.9 | — | — | — | |
| TESTRFPS=5.32025.02 | 73.3 | 83.9 | — | — | — | |
| MANGOYear=2021, Character-level annotations=false, Multi-scale=false2022.11 | 72.9 | 83.6 | 4.3 | — | — | |
| MANGOScale=L: 1600, Character-level annotations=true2022.07 | 72.9 | 83.6 | — | 4.3 | — | |
| MANGOExternal Data=Synth800K+Synth150K+COCO-Text+MLT19+IC13+IC15, Character-level Annotations=true, FPS (report)=4.32023.05 | 72.9 | 83.6 | — | — | — | |
| MANGOFPS=8.42025.02 | 72.9 | 83.6 | — | — | — | |
| Mask TextSpotter v3Year=2020, Character-level annotations=false, Multi-scale=false2022.11 | 71.2 | 78.4 | 2.5 | — | — | |
| Mask Textspotter v3Scale=S: 10002022.07 | 71.2 | 78.4 | — | — | — | |
| Mask TS v3External Data=Synth800K+IC13+IC15+SCUT2023.05 | 71.2 | 78.4 | — | — | — | |
| Mask TextSpotter v3FPS=3.02025.02 | 71.2 | 78.4 | — | — | — | |
| ABCNet v2Year=2021, Character-level annotations=false, Multi-scale=false2022.11 | 70.4 | 78.1 | 10 | — | — | |
| ABCNet v2External Data=Synth150K+MLT17, FPS (report)=10, FPS (A100)=14.92023.05 | 70.4 | 78.1 | — | — | — | |
| ABCNet v2FPS=10.02025.02 | 70.4 | 78.1 | — | — | — | |
| TESTRScale=L: 16002022.07 | 69.85 | 80.51 | — | 5.3 | 8.2 | |
| TextPerceptronScale=L: 13502022.07 | 69.7 | 78.3 | — | — | — | |
| TextPerceptronExternal Data=Synth800K2023.05 | 69.7 | 78.3 | — | — | — | |
| TextPerceptron2025.02 | 69.7 | 78.3 | — | — | — | |
| PAN++Year=2021, Character-level annotations=false, Multi-scale=false2022.11 | 68.6 | 78.6 | 21.1 | — | — | |
| PAN++External Data=Synth800K+COCO-Text+MLT17+IC15, FPS (report)=21.12023.05 | 68.6 | 78.6 | — | — | — | |
| PAN++FPS=36.02025.02 | 68.6 | 78.6 | — | — | — | |
| ABCNet v2Scale=S: 10002022.07 | 67.89 | 79.57 | — | 10 | 9.36 | |
| Qin et al.Year=2019, Character-level annotations=false, Multi-scale=false2022.11 | 67.8 | — | 4.8 | — | — | |
| Qin et al.Scale=S: 6002022.07 | 67.8 | — | — | 4.8 | — | |
| Mask TextSpotter'19Year=2019, Character-level annotations=false, Multi-scale=false2022.11 | 65.3 | 77.4 | 2 | — | — | |
| Mask Textspotter v2Scale=S: 10002022.07 | 65.3 | 77.4 | — | — | — | |
| BoundaryScale=L: 11002022.07 | 65 | 76.1 | — | — | — | |
| BoundaryExternal Data=Synth800K+IC13+IC152023.05 | 65 | 76.1 | — | — | — | |
| Boundary2025.02 | 65 | 76.1 | — | — | — | |
| PAN++Scale=S: 5122022.07 | 64.9 | 75.7 | — | 29.2 | 15.31 | |
| ABCNetYear=2020, Character-level annotations=false, Multi-scale=false2022.11 | 64.2 | 75.7 | 17.9 | — | — | |
| ABCNetScale=S: 10002022.07 | 63.74 | 77.62 | — | 17.9 | 14.59 | |
| CharNet H-57Character-level annotations=true2022.07 | 63.6 | — | — | — | — | |
| CharNetExternal Data=Synth800K, Character-level Annotations=true2023.05 | 63.6 | — | — | — | — | |
| CharNetFPS=5.42025.02 | 63.6 | — | — | — | — | |
| PGNetYear=2021, Character-level annotations=false, Multi-scale=false2022.11 | 63.1 | — | 35.5 | — | — | |
| PGNetExternal Data=Synth800K+IC15, FPS (report)=35.52023.05 | 63.1 | — | — | — | — | |
| PGNetFPS=38.22025.02 | 63.1 | — | — | — | — | |
| PGNet-AScale=L: 6402022.07 | 61.7 | — | — | 38.2 | 15.37 | |
| Feng et al.Year=2021, Character-level annotations=false, Multi-scale=false2022.11 | 55.8 | 79.2 | 7.2 | — | — | |
| TextNetScale=L: 9202022.07 | 54.02 | — | — | — | — | |
| TextDragonYear=2019, Character-level annotations=false, Multi-scale=false2022.11 | 48.8 | 74.8 | 2.6 | — | — | |
| TextDragon2022.07 | 48.8 | 74.8 | — | — | — | |
| TextDragonExternal Data=Synth800K2023.05 | 48.8 | 74.8 | — | — | — | |
| TextDragonFPS=2.62025.02 | 48.8 | 74.8 | — | — | — |