Startseite Neuigkeiten Neueste Forschungsarbeiten Tutorials Datensätze Wiki SOTA LLM-Modelle GPU-Rangliste Veranstaltungen

Deutsch

Scene Text Detection On Scut Ctw1500

Metriken

F-Measure

Precision

Recall

Ergebnisse

Leistungsergebnisse verschiedener Modelle zu diesem Benchmark

Modellname	F-Measure	Precision	Recall	Paper Title	Repository
PAN-640	83.7	86.4	81.2	Efficient and Accurate Arbitrary-Shaped Text Detection with Pixel Aggregation Network
FAST-S-512	82	85.6	78.7	FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation
SRFormer (ResNet-50)	89.6	91.6	87.7	SRFormer: Text Detection Transformer with Incorporated Segmentation and Regression
PSENet-1s	81.17	82.5	79.89	Shape Robust Text Detection with Progressive Scale Expansion Network
FAST-B-640	84.2	87.8	80.9	FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation
I3CL + SSL	86.5	88.4	84.6	I3CL:Intra- and Inter-Instance Collaborative Learning for Arbitrary-shaped Scene Text Detection
TextSnake	75.6	67.9	85.3	TextSnake: A Flexible Representation for Detecting Text of Arbitrary Shapes
FAST-T-512	81.5	85.5	77.9	FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation
SLPR	74.8	80.1	70.1	Sliding Line Point Regression for Shape Robust Scene Text Detection
FAST-B-512	82.9	85.7	80.2	FAST: Faster Arbitrarily-Shaped Text Detector with Minimalist Kernel Representation
DB-ResNet50 (1024)	83.4	-	-	Real-time Scene Text Detection with Differentiable Binarization
PAN	85	86.8	83.2	Mask R-CNN with Pyramid Attention Network for Scene Text Detection	-
MixNet	89.8	91.4	88.3	MixNet: Toward Accurate Detection of Challenging Scene Text in the Wild
PSENet-1s	82.2	84.8	79.7	Shape Robust Text Detection with Progressive Scale Expansion Network
TextFuseNet (ResNeXt-101)	87.4	89.7	85.1	TextFuseNet: Scene Text Detection with Richer Fused Features
DPText-DETR (ResNet50)	88.8	91.7	86.2	DPText-DETR: Towards Better Scene Text Detection with Dynamic Points in Transformer
CRAFT	83.5	86	81.1	Character Region Awareness for Text Detection

0 of 17 row(s) selected.