Text Spotting on ICDAR 2015

Metric: F-measure (%) - Strong Lexicon (higher is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	F-measure (%) - Strong Lexicon▼	Extra Data	Paper	Date↕	Code
1	UNITS	89	No	Towards Unified Scene Text Spotting based on Seq...	2023-04-07	Code
2	DeepSolo (ViTAEv2-S, TextOCR)	88.1	No	DeepSolo: Let Transformer Decoder with Explicit ...	2022-11-19	Code
3	DeepSolo(ResNet-50, TextOCR)	88	No	DeepSolo: Let Transformer Decoder with Explicit ...	2022-11-19	Code
4	DeepSolo(ResNet-50)	86.8	No	DeepSolo: Let Transformer Decoder with Explicit ...	2022-11-19	Code
5	SRTS	85.6	No	Single Shot Self-Reliant Scene Text Spotter by D...	2022-07-15	Code
6	TESTR	85.2	No	Text Spotting Transformers	2022-04-05	Code
7	A3S	84.8	No	A3S: Adversarial learning of semantic representa...	2023-02-21	-
8	GLASS	84.7	No	GLASS: Global to Local Attention for Scene-Text ...	2022-08-05	Code
9	SwinTextSpotter	83.9	No	SwinTextSpotter: Scene Text Spotting via Better ...	2022-03-19	Code
10	FOTS	83.6	No	FOTS: Fast Oriented Text Spotting with a Unified...	2018-01-05	Code
11	PGNet	83.3	No	PGNet: Real-time Arbitrarily-Shaped Text Spottin...	2021-04-12	Code
12	MaskTextSpotter v3	83.3	No	Mask TextSpotter v3: Segmentation Proposal Netwo...	2020-07-18	Code
13	ABCNet v2	82.7	No	ABCNet v2: Adaptive Bezier-Curve Network for Rea...	2021-05-08	Code
14	TextDragon	82.5	No	-	-	-
15	SPTS v2	82.3	No	SPTS v2: Single-Point Scene Text Spotting	2023-01-04	Code
16	MANGO	81.8	No	MANGO: A Mask Attention Guided One-Stage Scene T...	2020-12-08	Code
17	TextPerceptron	80.5	No	Text Perceptron: Towards End-to-End Arbitrary-Sh...	2020-02-17	Code
18	SPTS	77.5	No	SPTS: Single-Point Text Spotting	2021-12-15	Code

#1UNITSSOTA
89
F-measure (%) - Strong Lexicon· 2023-04-07
Towards Unified Scene Text Spotting based on Sequence Generation Code
#2DeepSolo (ViTAEv2-S, TextOCR)SOTA
88.1
F-measure (%) - Strong Lexicon· 2022-11-19
DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting Code
#3DeepSolo(ResNet-50, TextOCR)
88
F-measure (%) - Strong Lexicon· 2022-11-19
DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting Code
#4DeepSolo(ResNet-50)
86.8
F-measure (%) - Strong Lexicon· 2022-11-19
DeepSolo: Let Transformer Decoder with Explicit Points Solo for Text Spotting Code
#5SRTSSOTA
85.6
F-measure (%) - Strong Lexicon· 2022-07-15
Single Shot Self-Reliant Scene Text Spotter by Decoupled yet Collaborative Detection and Recognition Code
#6TESTRSOTA
85.2
F-measure (%) - Strong Lexicon· 2022-04-05
Text Spotting Transformers Code
#7A3S
84.8
F-measure (%) - Strong Lexicon· 2023-02-21
A3S: Adversarial learning of semantic representations for Scene-Text Spotting
#8GLASS
84.7
F-measure (%) - Strong Lexicon· 2022-08-05
GLASS: Global to Local Attention for Scene-Text Spotting Code
#9SwinTextSpotterSOTA
83.9
F-measure (%) - Strong Lexicon· 2022-03-19
SwinTextSpotter: Scene Text Spotting via Better Synergy between Text Detection and Text Recognition Code
#10FOTSSOTA
83.6
F-measure (%) - Strong Lexicon· 2018-01-05
FOTS: Fast Oriented Text Spotting with a Unified Network Code
#11PGNet
83.3
F-measure (%) - Strong Lexicon· 2021-04-12
PGNet: Real-time Arbitrarily-Shaped Text Spotting with Point Gathering Network Code
#12MaskTextSpotter v3
83.3
F-measure (%) - Strong Lexicon· 2020-07-18
Mask TextSpotter v3: Segmentation Proposal Network for Robust Scene Text Spotting Code
#13ABCNet v2
82.7
F-measure (%) - Strong Lexicon· 2021-05-08
ABCNet v2: Adaptive Bezier-Curve Network for Real-time End-to-end Text Spotting Code
#14TextDragon
82.5
F-measure (%) - Strong Lexicon
No paper
#15SPTS v2
82.3
F-measure (%) - Strong Lexicon· 2023-01-04
SPTS v2: Single-Point Scene Text Spotting Code
#16MANGO
81.8
F-measure (%) - Strong Lexicon· 2020-12-08
MANGO: A Mask Attention Guided One-Stage Scene Text Spotter Code
#17TextPerceptron
80.5
F-measure (%) - Strong Lexicon· 2020-02-17
Text Perceptron: Towards End-to-End Arbitrary-Shaped Text Spotting Code
#18SPTS
77.5
F-measure (%) - Strong Lexicon· 2021-12-15
SPTS: Single-Point Text Spotting Code