WeEdit: A Dataset, Benchmark and Glyph-Guided Framework for Text-centric Image Editing
Hui Zhang1,2,
Juntao Liu1,
Zongkai Liu1,3,
Liqiang Niu1,
Fandong Meng1,
Zuxuan Wu2,
and
Yu-Gang Jiang2
This repository hosts the WeEdit Benchmark, a standardized evaluation suite for text-centric image editing. It is part of the WeEdit project, which provides a systematic framework for… See the full description on the dataset page:
https://huggingface.co/datasets/HuiZhang0812/WeEdit_benchmark.