generative_inpainting
generative_inpainting copied to clipboard
JiahuiYu
DeepFill v1/v2 with Contextual Attention and Gated Convolution, CVPR 2018, and ICCV 2019 Oral
I have an image that is masked already so i don't have the original image with me but. How to run inference on that? Because in the test.py code it...
I have tried changing versions of tensor flow but the last problem that I am getting stuck is at the following. Total size of trainable weights: 0G 9M 548K 958B...
Traceback (most recent call last): File "train.py", line 7, in from inpaint_model import InpaintCAModel File "/home/myang47/generative_inpainting/inpaint_model.py", line 13, in from neuralgym.ops.gan_ops import gan_hinge_loss File "/home/myang47/.pyenv/versions/tf/lib/python3.6/site-packages/neuralgym/ops/gan_ops.py", line 138, in class Conv2DSepctralNorm(tf.layers.Conv2D):...
Hi, sorry for bothering you. I was looking at your code and I noticed that `xconv2_downsample` and `xconv4_downsample` and `pmconv2_downsample` have a different number of channels wrt the supmat (/2)...
I have encountered `404: sorry, the page is here` when visiting the [demo](https://github.com/JiahuiYu/generative_inpainting), could you please fix it?
I have simply optimized the images by compressing their size without compromising their quality to make them more space-efficient
Hello, from the readme I see that "all the pretrained models are trained with images of resolution 256x256 and largest hole size 128x128". However, as this work is often used...
I want to know whether i can use my own random masks for training,thanks
num_gpus_per_job: 1 # number of gpus each job need num_cpus_per_job: 4 # number of gpus each job need num_hosts_per_job: 1 memory_per_job: 32 # number of gpus each job need gpu_type:...
Hi， I want to use my own dataset for training, but instead of using a generative-adversarial network, I prefer only use the generator part to generate predictions and calculate the...
