synth_doc_generation
synth_doc_generation copied to clipboard
Official PyTorch Implementation of DocSynth: A Layout Guided Approach for Controllable Document Image Synthesis - ICDAR 2021
Results
3
synth_doc_generation issues
Sort by
recently updated
recently updated
newest added
I want to train my model for word detection in documents. Is there anyway I could get word level annotations for these images.
The publaynet_netG.pkl checkpoint provided in the repository is not generating desired meaningful outputs. I tried running `test.py` using the pretrained ckpt but the output generations for the test data(publaynet dataset)...
Hi, thank you very much for this code. I have only one question: how may we generate 128x128 images instead 64x64 using `test.py`? It is possible to use the pretrained...