luanfujun / deep-painterly-harmonization

Code and data for paper "Deep Painterly Harmonization": https://arxiv.org/abs/1804.03189
6.08k stars 628 forks source link

Does it handle large images? #13

Open jayavanth opened 6 years ago

jayavanth commented 6 years ago

I tried training it with a 1440 image and it throws an out of memory error. I did use --image_size 1440

luanfujun commented 6 years ago

Sorry but it can't work with such large resolution images due to GPU memory limitation. It could be modified to do it by making use of multiple GPUs.

DrugRui commented 4 years ago

Sorry but it can't work with such large resolution images due to GPU memory limitation. It could be modified to do it by making use of multiple GPUs.

Hello, my image size is 960*540. After I got the results,I encountered some problems in post-processing:

Octave: 2 > run (' filt_cnn_artifact. M ')
Invalid use of script/home/deep - painterly - harmonization / 3 rdparty/patchmatch - 2.0 / nnmex. M in index expression
Error: called the from
Filt_cnn_artifact at line 50 column 9
Run at line 86 column 5

Do I need to change some parameters of patchmatch?

tombohub commented 4 years ago

@DrugRui hello, did you find solution for that error?