Scene Text Recognition on Street View Text unconstrained (test)
87AccuracyDICT-SVT-Full
Evaluation results
| Method | Links | |
|---|---|---|
| DICT-SVT-FullLanguage Model=Tightly constrained (SVT-Full)2014.06 | 87 | |
| DICT+2-90kLexicon=90k-word2014.06 | 80.7 | |
| PhotoOCR2014.06 | 78 | |
| CHAR+2Language Model=None2014.06 | 68 |