HyperAI超神经

首页资讯最新论文教程数据集百科 SOTA LLM 模型天梯 GPU 天梯顶会

中文

HyperAI超神经

Scene Text Detection On Icdar 2017 Mlt 1

评估指标

F-Measure

Precision

Recall

评测结果

各个模型在此基准测试上的表现结果

模型名称	F-Measure	Precision	Recall	Paper Title	Repository
FOTS MS	70.75%	81.86	62.3	FOTS: Fast Oriented Text Spotting with a Unified Network
GNNets	74.54%	79.63	70.06	Geometry Normalization Networks for Accurate Scene Text Detection
CharNet R-50	73.42%	77.07	70.1	Convolutional Character Networks
PMTD*	80.13%	84.42	76.25	Pyramid Mask Text Detector
FOTS	67.25%	80.95	57.51	FOTS: Fast Oriented Text Spotting with a Unified Network
SBD	79.47%	82.75	76.44	Exploring the Capacity of an Orderless Box Discretization Network for Multi-orientation Scene Text Detection
CharNet H-88	75.77%	81.27	70.97	Convolutional Character Networks
CRAFT	-	80.6	68.2	Character Region Awareness for Text Detection
PAN	74.3%	80	69.8	Mask R-CNN with Pyramid Attention Network for Scene Text Detection	-
Corner Localization (single-scale)	66.8%	83.8	55.6	Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation
SPCNET	74.1%	80.6	68.6	Scene Text Detection with Supervised Pyramid Context Network
PSENet (ResNet-152)	72.13%	75.35	69.18	Shape Robust Text Detection with Progressive Scale Expansion Network
Corner Localization (multi-scale)	72.4%	74.3	70.6	Multi-Oriented Scene Text Detection via Corner Localization and Region Segmentation
PSENet-1s	72.45%	77.01	68.4	Shape Robust Text Detection with Progressive Scale Expansion Network

0 of 14 row(s) selected.