Mask TextSpotter: An End-to-End Trainable Neural Network for Spotting Text with Arbitrary Shapes

Recently, models based on deep neural networks have dominated the fields of scene text detection and recognition. In this paper, we investigate the problem of scene text spotting, which aims at simultaneous text detection and recognition in natural images... (read more)

PDF Abstract

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Scene Text Detection ICDAR 2013 Mask TextSpotter F-Measure 91.7% # 3
Precision 95 # 3
Recall 88.6 # 4
Scene Text Detection ICDAR 2015 Mask TextSpotter F-Measure 86 # 17
Precision 91.6 # 6
Recall 81 # 22
Scene Text Detection Total-Text Mask TextSpotter F-Measure 61.3% # 13
Precision 69 # 11
Recall 55 # 10

Methods used in the Paper


METHOD TYPE
Softmax
Output Functions
RoIAlign
RoI Feature Extractors
Convolution
Convolutions
Mask R-CNN
Instance Segmentation Models