NRTR+TPS++

Reported on 6 benchmarks across 3 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Computer Vision4 results

Scene ParsingonSVT
Accuracy· 2023-05-09
94.6
best: 99.1 (CLIP4STR-H (DFN-5B))
TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition arXiv:2305.05322
Scene ParsingonCUTE80
Accuracy· 2023-05-09
92.4
best: 99.7 (CPPD)
TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition arXiv:2305.05322
Scene Text RecognitiononSVT
Accuracy· 2023-05-09
94.6
best: 99.1 (CLIP4STR-H (DFN-5B))
TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition arXiv:2305.05322
Scene Text RecognitiononCUTE80
Accuracy· 2023-05-09
92.4
best: 99.7 (CPPD)
TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition arXiv:2305.05322

Audio2 results

2D Semantic SegmentationonSVT
Accuracy· 2023-05-09
94.6
best: 99.1 (CLIP4STR-H (DFN-5B))
TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition arXiv:2305.05322
2D Semantic SegmentationonCUTE80
Accuracy· 2023-05-09
92.4
best: 99.7 (CPPD)
TPS++: Attention-Enhanced Thin-Plate Spline for Scene Text Recognition arXiv:2305.05322