ML Research Wiki / Benchmarks / Scene Text Recognition / SVTP

SVTP

Scene Text Recognition Benchmark

Performance Over Time

📊 Showing 16 results | 📏 Metric: Accuracy

Top Performing Models

Rank Model Paper Accuracy Date Code
1 DTrOCR 105M DTrOCR: Decoder-only Transformer for Optical Character Recognition 98.60 2023-08-30 📦 arvindrajan92/DTrOCR
2 MGP-STR 📚 Multi-Granularity Prediction for Scene Text Recognition 98.30 2022-09-08 📦 alibabaresearch/advancedliteratemachinery 📦 AlibabaResearch/AdvancedLiterateMachinery 📦 topdu/openocr
3 CLIP4STR-L* 📚 An Empirical Study of Scaling Law for OCR 98.13 2023-12-29 📦 large-ocr-model/large-ocr-model.github.io
4 CLIP4STR-L (DataComp-1B) CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model 98.10 2023-05-23 📦 VamosC/CLIP4STR
5 CLIP4STR-L CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model 97.40 2023-05-23 📦 VamosC/CLIP4STR
6 CLIP4STR-B 📚 CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model 97.20 2023-05-23 📦 VamosC/CLIP4STR
7 CPPD 📚 Context Perception Parallel Decoder for Scene Text Recognition 96.70 2023-07-23 📦 PaddlePaddle/PaddleOCR 📦 topdu/openocr
8 CCD-ViT-Base 📚 Self-supervised Character-to-Character Distillation for Text Recognition 96.10 2022-11-01 📦 tongkunguan/ccd
9 CCD-ViT-Small 📚 Self-supervised Character-to-Character Distillation for Text Recognition 92.70 2022-11-01 📦 tongkunguan/ccd
10 CCD-ViT-Tiny 📚 Self-supervised Character-to-Character Distillation for Text Recognition 91.60 2022-11-01 📦 tongkunguan/ccd

All Papers (16)

arvindrajan92/DTrOCR
alibabaresearch/advancedliteratemachinery AlibabaResearch/AdvancedLiterateMachinery topdu/openocr
large-ocr-model/large-ocr-model.github.io
VamosC/CLIP4STR
VamosC/CLIP4STR
VamosC/CLIP4STR
PaddlePaddle/PaddleOCR topdu/openocr
tongkunguan/ccd
tongkunguan/ccd
tongkunguan/ccd
adeline-cs/GTR
topdu/openocr byeonghu-na/matrn wp03052/MATRN
tongkunguan/siga
topdu/openocr simplify23/CDistNet chibohe/CdistNet-pytorch
topdu/openocr baudm/parseq