这款应用很强大但demo是由js动态生成html的标签,所以标签修改容易,需要对js的相关接口很熟悉,同时需要下载50m 左右 的weight权重
源码是挂在pc端网页上的,但现已经全部改成h5移动端的形式
以下是些在比较重要的配置,可以略过
pix2pix(lua )官方—github地址:在这里可以下载相关的权重
bash ./models/download_model.sh model_name
- facades_label2image (label -> facade): trained on the CMP Facades dataset.
- cityscapes_label2image (label -> street scene): trained on the Cityscapes dataset.
- cityscapes_image2label (street scene -> label): trained on the Cityscapes dataset.
- edges2shoes (edge -> photo): trained on UT Zappos50K dataset.
- edges2handbags (edge -> photo): trained on Amazon handbags images.
- day2night (daytime scene -> nighttime scene): trained on around 100 webcams.
pix2pix(tensorflow-js)—github地址:
demo
pix2pix-tensorflow-models
# clone this repo
git clone https://github.com/affinelayer/pix2pix-tensorflow.git
cd pix2pix-tensorflow
#下载CMP Facades的数据集(从http://cmp.felk.cvut.cz/~tylecr1/facade/)
python tools/download-dataset.py facades
#训练模型(这可能花1到8小时取决于GPU,cpu将会更久)
python pix2pix.py \
--mode train \
--output_dir facades_train \
--max_epochs 200 \
--input_dir facades/train \
--which_direction BtoA
test the model
--mode test \
--output_dir facades_test \
--input_dir facades/val \
--checkpoint facades_train```
http://efrosgans.eecs.berkeley.edu/pix2pix/models-pytorch/
钢笔绘画效果