From Two to One: A New Scene Text Recognizer with Visual Language Modeling Network

Benchmark Model Rank Results
scene-text-recognition-on-hostVisionLAN (extra data)#61:1 Accuracy: 50.3
scene-text-recognition-on-wostVisionLAN (extra data)#91:1 Accuracy: 70.3
scene-text-recognition-on-iiit5kVisionLAN (extra data)#37Accuracy: 95.8
scene-text-recognition-on-svtpVisionLAN (extra data)#37Accuracy: 86.0
scene-text-recognition-on-cute80VisionLAN (extra data)#40Accuracy: 88.5
scene-text-recognition-on-svtVisionLAN (extra data)#41Accuracy: 91.7