SOTAVerified

Text Spotting

Scene Text Spotting is the combination of Scene Text Detection and Scene Text Recognition in an end-to-end manner. It is the ability to read natural text in the wild.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 1–10 of 112 papers

TitleStatusHype
Text-Aware Image Restoration with Diffusion Models—0
GoMatching++: Parameter- and Data-Efficient Arbitrary-Shaped Video Text Spotting and BenchmarkingCode1
SemiETS: Integrating Spatial and Content Consistencies for Semi-Supervised End-to-end Text SpottingCode1
TextInPlace: Indoor Visual Place Recognition in Repetitive Structures with Scene Text Spotting and VerificationCode1
OmniParser V2: Structured-Points-of-Thought for Unified Visual Text Parsing and Its Generality to Multimodal Large Language ModelsCode0
CLIP is Almost All You Need: Towards Parameter-Efficient Scene Text Retrieval without OCR—0
Hear the Scene: Audio-Enhanced Text Spotting—0
InstructOCR: Instruction Boosting Scene Text SpottingCode0
Arbitrary Reading Order Scene Text Spotter with Local Semantics Guidance—0
HIP: Hierarchical Point Modeling and Pre-training for Visual Information Extraction—0
Show:102550
← PrevPage 1 of 12Next →
All datasetsICDAR 2015Total-TextSCUT-CTW1500Inverse-Text

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1DeepSolo (ViTAEv2-S, TextOCR)F-measure (%) - No Lexicon68.8—Unverified
2DeepSolo (ResNet-50, TextOCR)F-measure (%) - No Lexicon64.6—Unverified
3SwinTextSpotterF-measure (%) - No Lexicon55.4—Unverified
4DeepSolo (ResNet-50)F-measure (%) - No Lexicon48.5—Unverified
5MaskTextSpotter v2F-measure (%) - No Lexicon39—Unverified
6SPTSF-measure (%) - No Lexicon38.3—Unverified
7ABCNet v2F-measure (%) - No Lexicon34.5—Unverified
8TESTRF-measure (%) - No Lexicon34.2—Unverified
9ABCNetF-measure (%) - No Lexicon22.2—Unverified