This is the implementation of the paper Interactive Sketch-Based Normal Map Generation with Deep Neural Networks
We run the program on a Linux desktop using python.
Train the model:
pyhton main.py --phase train --dataset_name <dataset>
Test:
python main.py --phase test --dataset_name <dataset>
Visualize the training process:
cd logs
tensorboard --logdir=./
Data: Training and testing data can be downloaded here. After extracting the compressed file, put the folder (datasets) in the project directory (the same directory where the main file locates in).
This code is based on the implementation of pix2pix from pix2pix-tensorflow. Thanks for the great work!