HyperAI
StartseiteNeuigkeitenNeueste ForschungsarbeitenTutorialsDatensätzeWikiSOTALLM-ModelleGPU-RanglisteVeranstaltungen
Suche
Über
HyperAI

Scene Text Detection On Icdar 2013

Metriken

F-Measure
Precision
Recall

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname
F-Measure
Precision
Recall
Paper TitleRepository
SPCNET92.1%93.890.5Scene Text Detection with Supervised Pyramid Context Network
PixelLink+VGG16 2s MS88.1%88.687.5PixelLink: Detecting Scene Text via Instance Segmentation
Gupta et al.83.0%92.075.5Synthetic Data for Text Localisation in Natural Images
Neumann et al. *77.1%81.872.4Efficient Scene Text Localization and Recognition with Local Character Refinement-
SSTD87%8886Single Shot Text Detector with Regional Attention
USM (COCO TS + ICDAR–2013)80.40%--Unsharp Masking Layer: Injecting Prior Knowledge in Convolutional Networks for Image Classification
Corner Localization (multi-scale)88%9284.4Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation
CRAFT-97.493.1Character Region Awareness for Text Detection
SegLink85.3%87.783Detecting Oriented Text in Natural Images by Linking Segments
STN-OCR90.3%--STN-OCR: A single Neural Network for Text Detection and Text Recognition
WordSup (VGG16-synth-icdar)90.34%93.3487.53WordSup: Exploiting Word Annotations for Character based Text Detection-
TextBoxes++_MS88%%9184TextBoxes++: A Single-Shot Oriented Scene Text Detector
TextFuseNet (ResNeXt-101)94.61%97.2792.09TextFuseNet: Scene Text Detection with Richer Fused Features
Jaderberg et al.76.8%88.567.8Reading Text in the Wild with Convolutional Neural Networks-
Corner-based Region Proposals87.6%%91.983.9Detecting Multi-Oriented Text with Corner-based Region Proposals
Mask TextSpotter91.7%9588.6Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes
0 of 16 row(s) selected.