OCRVQA
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Visual Question Answering | OCRVQA | Accuracy54.9 | 47 | |
| Scene Text-Centric Visual Question Answering | OCRVQA | Accuracy64.4 | 14 | |
| OCR-based Visual Question Answering | OCRVQA | Mean Accuracy63.2 | 13 | |
| OCR-based Visual Question Answering | OCRVQA 2019 (test) | Accuracy61.4 | 13 |
Showing 4 of 4 rows