Scene Text Recognition2020en

uber-text

Dataset from Papers With Code

Metrics:accuracy, cer, wer, f1
Current State of the Art

CLIP4STR-L (DataComp-1B)

Unknown

92.2

accuracy

accuracy Progress Over Time

Showing 2 breakthroughs from Sep 2022 to May 2023

90.991.291.692.092.3Sep 2022May 2023accuracyDate

Key Milestones

Sep 2022
MGP-STR

From paper: Multi-Granularity Prediction for Scene Text Recognition

91.0
Source
May 2023
CLIP4STR-L (DataComp-1B)Current SOTA

From paper: CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model

92.2
+1.3%
Source
Total Improvement
1.3%
Time Span
8m
Breakthroughs
2
Current SOTA
92.2

Top Models Performance Comparison

Top 3 models ranked by accuracy

accuracy1CLIP4STR-L (DataComp-1B)92.2100.0%2MGP-STR91.098.7%3CLIP4STR-B*86.894.1%0%25%50%75%100%% of best
Best Score
92.2
Top Model
CLIP4STR-L (DataC...
Models Compared
3
Score Range
5.4

accuracyPrimary

#ModelScorePaper / CodeDate
1
CLIP4STR-L (DataComp-1B)
92.2
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language ModelCode
May 2023
2
MGP-STR
91
Multi-Granularity Prediction for Scene Text RecognitionCode
Sep 2022
3
CLIP4STR-B*
86.8
CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language ModelCode
May 2023

Related Papers2

CLIP4STR: A Simple Baseline for Scene Text Recognition with Pre-trained Vision-Language Model
May 2023Models: CLIP4STR-L (DataComp-1B), CLIP4STR-B*
PaperCode
Multi-Granularity Prediction for Scene Text Recognition
Sep 2022Models: MGP-STR
PaperCode

Other Scene Text Recognition Datasets

svtic13svtpcute80wostsvt-phostmsdaicdar-2003iiit5k
uber-text Benchmark - Scene Text Recognition | CodeSOTA