Skip to main content

Table 2 Results on ICDAR 2015

From: An end-to-end text spotter with text relation networks

Method Detection End-to-End
  P R F S W G
EAST(Zhou et al. 2017) 83.3 78.3 80.7 - - -
TextBoxes++(Liao et al. 2018) 87.2 76.7 81.7 73.3 65.9 51.9
He et al.(He et al. 2018) 87.0 86.0 87.0 82.0 77.0 63.0
TextNet(Sun et al. 2018) 89.4 85.4 87.4 78.7 74.9 60.5
FOTS(Liu et al. 2018) 91.0 85.2 88.0 80.1 75.9 60.8
Mask TextSpotter(Lyu et al. 2018) 91.6 81.0 86.0 79.3 73.0 62.4
det-baseline 93.3 75.2 83.3 - - -
e2e-baseline 94.2 76.7 84.6 75.6 69.6 61.0
model with TRN 95.2 76.1 84.6 78.7 74.2 67.5
  1. The entries in boldface mean that the best performances for different measurements are obtained by compared works