HyperAI超神经

首页资讯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

Table Recognition On Pubtabnet

评估指标

TEDS (all samples)

TEDS-Struct

评测结果

各个模型在此基准测试上的表现结果

模型名称	TEDS (all samples)	TEDS-Struct	Paper Title	Repository
TRUST	96.2	97.1	TRUST: An Accurate and End-to-End Table structure Recognizer Using Splitting-based Transformers	-
ConvStem	96.53	-	High-Performance Transformers for Table Structure Recognition Need Early Convolutions
EDD	88.3	-	Image-based table recognition: data, model, and evaluation
LGPMA	94.6	96.7	LGPMA: Complicated Table Structure Recognition with Local and Global Pyramid Mask Alignment
TabStruct-Net	90.1	90.1	Table Structure Recognition using Top-Down and Bottom-Up Cues
MuTabNet	96.87	-	Multi-Cell Decoder and Mutual Learning for Table Structure and Character Recognition
TableMaster	96.76	-	PingAn-VCGroup's Solution for ICDAR 2021 Competition on Scientific Literature Parsing Task B: Table Recognition to HTML
SEM	93.7	-	Split, embed and merge: An accurate table structure recognizer	-
TSRFormer	-	97.5	TSRFormer: Table Structure Recognition with Transformers	-
SLANet	96.3	97.01	PP-StructureV2: A Stronger Document Analysis System
RTSR	-	97	Robust Table Detection and Structure Recognition from Heterogeneous Document Images	-
NCGM	95.4	-	Neural Collaborative Graph Machines for Table Structure Recognition	-
Multi-Task Learning Model	96.67	97.88	An End-to-End Multi-Task Learning Model for Image-based Table Recognition

0 of 13 row(s) selected.