WordSup: Exploiting Word Annotations for Character based Text Detection

ICCV 2017 Han HuChengquan ZhangYuxuan LuoYuzhuo WangJunyu HanErrui Ding

Imagery texts are usually organized as a hierarchy of several visual elements, i.e. characters, words, text lines and text blocks. Among these elements, character is the most basic one for various languages such as Western, Chinese, Japanese, mathematical expression and etc... (read more)

PDF Abstract

Code


No code implementations yet. Submit your code now

Results from the Paper


TASK DATASET MODEL METRIC NAME METRIC VALUE GLOBAL RANK RESULT BENCHMARK
Scene Text Detection COCO-Text WordSup (VGG16-synth-coco) F-Measure 36.8 # 5
Precision 45.2 # 5
Recall 30.9 # 5
Scene Text Detection ICDAR 2013 WordSup (VGG16-synth-icdar) F-Measure 90.34% # 3
Precision 93.34 # 4
Recall 87.53 # 4
Scene Text Detection ICDAR 2015 WordSup (VGG16-synth-icdar) F-Measure 78.2 # 24
Precision 79.3 # 27
Recall 77.0 # 28

Methods used in the Paper