SOTAVerified

Scene Text Detection

Scene Text Detection is a computer vision task that involves automatically identifying and localizing text within natural images or videos. The goal of scene text detection is to develop algorithms that can robustly detect and and label text with bounding boxes in uncontrolled and complex environments, such as street signs, billboards, or license plates.

Source: ContourNet: Taking a Further Step toward Accurate Arbitrary-shaped Scene Text Detection

Papers

Showing 5175 of 213 papers

TitleStatusHype
Sliding Line Point Regression for Shape Robust Scene Text DetectionCode0
COCO-Text: Dataset and Benchmark for Text Detection and Recognition in Natural ImagesCode0
Selective Style Transfer for TextCode0
STELA: A Real-Time Scene Text Detector with Learned AnchorCode0
DocXChain: A Powerful Open-Source Toolchain for Document Parsing and BeyondCode0
Scene Text Detection with Supervised Pyramid Context NetworkCode0
Robust Scene Text Recognition with Automatic RectificationCode0
Scene Text Detection and Recognition: The Deep Learning EraCode0
Detecting Text in Natural Image with Connectionist Text Proposal NetworkCode0
Detecting Oriented Text in Natural Images by Linking SegmentsCode0
Detecting Curve Text in the Wild: New Dataset and New SolutionCode0
Single Shot Text Detector with Regional AttentionCode0
SEE: Towards Semi-Supervised End-to-End Scene Text RecognitionCode0
STEP -- Towards Structured Scene-Text SpottingCode0
Rethinking Irregular Scene Text RecognitionCode0
Pyramid Mask Text DetectorCode0
Progressive Contour Regression for Arbitrary-Shape Scene Text DetectionCode0
R2CNN: Rotational Region CNN for Orientation Robust Scene Text DetectionCode0
PGNet: Real-time Arbitrarily-Shaped Text Spotting with Point Gathering NetworkCode0
Accurate Text Localization in Natural Image with Cascaded Convolutional Text NetworkCode0
PixelLink: Detecting Scene Text via Instance SegmentationCode0
Dataset and Benchmark for Urdu Natural Scenes Text Detection, Recognition and Visual Question AnsweringCode0
A Single-Shot Arbitrarily-Shaped Text Detector based on Context Attended Multi-Task LearningCode0
Multi-Oriented Scene Text Detection via Corner Localization and Region SegmentationCode0
Multi-Oriented Text Detection with Fully Convolutional NetworksCode0
Show:102550
← PrevPage 3 of 9Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1TextFuseNet (ResNeXt-101)F-Measure92.23Unverified
2CharNet H-88 (multi-scale)F-Measure91.55Unverified
3CharNet H-88 (single-scale)F-Measure90.97Unverified
4CharNet H-50 (multi-scale)F-Measure90.16Unverified
5SBDF-Measure90.1Unverified
6CharNet H-57 (multi-scale)F-Measure90.06Unverified
7FOTS MSF-Measure89.84Unverified
8CharNet H-50 (single-scale)F-Measure89.7Unverified
9CharNet H-57 (single-scale)F-Measure89.66Unverified
10PMTDF-Measure89.33Unverified
#ModelMetricClaimedVerifiedStatus
1MixNetF-Measure90.5Unverified
2SRFormer (ResNet-50)F-Measure90Unverified
3DPText-DETR (ResNet-50)F-Measure89Unverified
4TextFuseNet (ResNeXt-101)F-Measure87.5Unverified
5FAST-B-800F-Measure87.5Unverified
6I3CL + SSL(ResNet-50)F-Measure86.9Unverified
7CharNet H-88 (multi-scale)F-Measure86.5Unverified
8FAST-B-640F-Measure86.4Unverified
9DBNet++ (ResNet-50) (800)F-Measure86Unverified
10FAST-B-512F-Measure85.8Unverified
#ModelMetricClaimedVerifiedStatus
1MixNetF-Measure89.4Unverified
2FAST-B-736F-Measure87.3Unverified
3DBNet++ (ResNet-50) (736)F-Measure87.2Unverified
4FAST-S-736F-Measure86.4Unverified
5DBNet++ (ResNet-18) (736)F-Measure85.1Unverified
6DB-ResNet-50 (736)F-Measure84.9Unverified
7FAST-T-736F-Measure84.9Unverified
8FAST-T-512F-Measure84.5Unverified
9PANF-Measure84.1Unverified
10CRAFTF-Measure82.9Unverified
#ModelMetricClaimedVerifiedStatus
1MixNetF-Measure89.8Unverified
2SRFormer (ResNet-50)F-Measure89.6Unverified
3DPText-DETR (ResNet50)F-Measure88.8Unverified
4TextFuseNet (ResNeXt-101)F-Measure87.4Unverified
5I3CL + SSLF-Measure86.5Unverified
6PANF-Measure85Unverified
7FAST-B-640F-Measure84.2Unverified
8PAN-640F-Measure83.7Unverified
9CRAFTF-Measure83.5Unverified
10DB-ResNet50 (1024)F-Measure83.4Unverified
#ModelMetricClaimedVerifiedStatus
1CRAFTPrecision97.4Unverified
2TextFuseNet (ResNeXt-101)F-Measure94.61Unverified
3SPCNETF-Measure92.1Unverified
4Mask TextSpotterF-Measure91.7Unverified
5WordSup (VGG16-synth-icdar)F-Measure90.34Unverified
6STN-OCRF-Measure90.3Unverified
7PixelLink+VGG16 2s MSF-Measure88.1Unverified
8TextBoxes++_MSF-Measure88Unverified
9Corner Localization (multi-scale)F-Measure88Unverified
10Corner-based Region ProposalsF-Measure87.6Unverified
#ModelMetricClaimedVerifiedStatus
1PMTD*Precision84.42Unverified
2Corner Localization (single-scale)Precision83.8Unverified
3SBDPrecision82.75Unverified
4FOTS MSPrecision81.86Unverified
5CharNet H-88Precision81.27Unverified
6FOTSPrecision80.95Unverified
7CRAFTPrecision80.6Unverified
8SPCNETPrecision80.6Unverified
9PANPrecision80Unverified
10GNNetsPrecision79.63Unverified
#ModelMetricClaimedVerifiedStatus
1Corner-based Region ProposalsF-Measure59.1Unverified
2TextBoxes++_MSF-Measure58.72Unverified
3EAST + VGG16F-Measure39.45Unverified
4SSTDF-Measure37Unverified
5WordSup (VGG16-synth-coco)F-Measure36.8Unverified
6Yao et al.F-Measure33.31Unverified
#ModelMetricClaimedVerifiedStatus
1MixNetH-Mean79.7Unverified
2SRFormer (ResNet-50)H-Mean79.3Unverified
3TextFuseNet (ResNeXt-101)H-Mean78.6Unverified
4DPText-DETR (ResNet-50)H-Mean78.1Unverified
#ModelMetricClaimedVerifiedStatus
1BDNF-Measure93.36Unverified