Codesota · Computer Vision · Scene Text Detection · coco-textTasks/Computer Vision/Scene Text Detection
Scene Text Detection · benchmark dataset · 2020 · EN

coco-text.

Dataset from Papers With Code

Submit a result
§ 01 · Leaderboard

Best published scores.

33 results indexed across 4 metrics. Shaded row marks current SOTA; ties broken by submission date.


Primary
accuracy · higher is better
All metrics
1-1-accuracy, f-measure, precision, recall
1-1-accuracy
3 rows
#ModelOrgSubmittedPaper / code1-1-accuracy
01CLIP4STR-LMay 2023CLIP4STR: A Simple Baseline for Scene Text Recognition w… · code81.90
02MGP-STRSep 2022Multi-Granularity Prediction for Scene Text Recognition · code81.70
03CLIP4STR-BResearchMay 2023CLIP4STR: A Simple Baseline for Scene Text Recognition w… · code81.10
f-measure
18 rows
#ModelOrgSubmittedPaper / codef-measure
01TCMOSSCLIP-basedMar 2026ijcai-2025-benchmark65.90
02PANet (Joint)OSSICCV 2019Mar 2026ijcai-2025-benchmark64.50
03LRANetOSSAAAI 2024Mar 2026ijcai-2025-benchmark61.70
04DPText-DETROSSAAAI 2023Mar 2026ijcai-2025-benchmark61.60
05MAEDetOSSIJCAI 2025Mar 2026ijcai-2025-benchmark60.60
06DBNetOSSAAAI 2020Mar 2026ijcai-2025-benchmark60.50
07DBNet++OSSTPAMI 2022Mar 2026ijcai-2025-benchmark59.50
08SRFormerOSSAAAI 2024Mar 2026ijcai-2025-benchmark59.40
09Corner-based Region ProposalsApr 2018Detecting Multi-Oriented Text with Corner-based Region P… · code59.10
10TextBoxes++_MSJan 2018TextBoxes++: A Single-Shot Oriented Scene Text Detector · code58.72
11FCENetOSSCVPR 2021Mar 2026ijcai-2025-benchmark57.90
12PSENetOSSCVPR 2019Mar 2026ijcai-2025-benchmark56
13ABCNet v2OSSTPAMI 2021Mar 2026ijcai-2025-benchmark53.20
14EAST + VGG16Apr 2017EAST: An Efficient and Accurate Scene Text Detector · code39.45
15SSTDSep 2017Single Shot Text Detector with Regional Attention · code37
16WordSup (VGG16-synth-coco)Aug 2017WordSup: Exploiting Word Annotations for Character based…36.80
17Yao et al.Jun 2016Scene Text Detection via Holistic, Multi-Channel Predict…33.31
18DRRGOSSCVPR 2020Mar 2026ijcai-2025-benchmark31.90
precision
6 rows
#ModelOrgSubmittedPaper / codeprecision
01TextBoxes++_MSJan 2018TextBoxes++: A Single-Shot Oriented Scene Text Detector · code60.87
02Corner-based Region ProposalsApr 2018Detecting Multi-Oriented Text with Corner-based Region P… · code55.50
03EAST + VGG16Apr 2017EAST: An Efficient and Accurate Scene Text Detector · code50.39
04SSTDSep 2017Single Shot Text Detector with Regional Attention · code46
05WordSup (VGG16-synth-coco)Aug 2017WordSup: Exploiting Word Annotations for Character based…45.20
06Yao et al.Jun 2016Scene Text Detection via Holistic, Multi-Channel Predict…43.23
recall
6 rows
#ModelOrgSubmittedPaper / coderecall
01Corner-based Region ProposalsApr 2018Detecting Multi-Oriented Text with Corner-based Region P… · code63.30
02TextBoxes++_MSJan 2018TextBoxes++: A Single-Shot Oriented Scene Text Detector · code56.70
03EAST + VGG16Apr 2017EAST: An Efficient and Accurate Scene Text Detector · code32.40
04SSTDSep 2017Single Shot Text Detector with Regional Attention · code31
05WordSup (VGG16-synth-coco)Aug 2017WordSup: Exploiting Word Annotations for Character based…30.90
06Yao et al.Jun 2016Scene Text Detection via Holistic, Multi-Channel Predict…27.10
Fig 2 · Rows sorted by score within each metric. Shaded row marks SOTA. Dates reflect model or paper release where available, otherwise the date Codesota accessed the source.
§ 04 · Literature

8 papers
tied to this benchmark.

Every paper below corresponds to at least one row in the leaderboard above. Click through for the arXiv preprint and, when available, the reference implementation.

§ 06 · Contribute

Have a score that beats
this table?

Submit a checkpoint and a reproduction script. We will run it, publish the score, and — if it takes the top — annotate the step on the progress chart with your name.

Submit a result Read submission guide
What a submission needs
  • 01A public checkpoint or API endpoint
  • 02A reproduction script with frozen commit + seed
  • 03Declared evaluation environment (Python, deps)
  • 04One row per metric declared by this dataset
  • 05A contact so we can follow up on discrepancies