ViTEraser: Harnessing the Power of Vision Transformers for Scene Text Removal with SegMIM Pretraining.

Dezhi Peng, Chongyu Liu, Yuliang Liu, Lianwen Jin

Browse the full AAAI paper archive.