depth-preserving-neural-style-transfe

This is a course project done by Ruizhi Liao, Yu Xia and Xiuming Zhang at MIT, within a very short time frame.

The report is especially unpolished, not even proof-read.

Please use the code with caution.

Name	Name	Last commit message	Last commit date
Latest commit aluex minor Jan 5, 2017 01bea5c · Jan 5, 2017 History 160 Commits
doc	doc	Update flags.md	Oct 6, 2016
fast_neural_style	fast_neural_style	Delete a wrong line	Dec 6, 2016
images	images	Restoring the size; it's not the issue	Dec 7, 2016
models	models	tweaking	Oct 4, 2016
relative-depth	relative-depth	Add relative-depth	Dec 3, 2016
report	report	minor	Jan 5, 2017
scripts	scripts	allow convert_network.lua to work when file has a .model field	Oct 26, 2016
test	test	Fix module name	Oct 13, 2016
.gitignore	.gitignore	update	Dec 13, 2016
README.md	README.md	Change the readme.md	Jan 5, 2017
fast_neural_style.lua	fast_neural_style.lua	Pass a bool to setup_gpu for use_cudnn	Oct 13, 2016
print_options.lua	print_options.lua	add script to print options for checkpoints	Oct 4, 2016
requirements.txt	requirements.txt	initial commit	Oct 4, 2016
slow_neural_style.lua	slow_neural_style.lua	Pass bool to setup_gpu at other call sites	Oct 13, 2016
train.lua	train.lua	Add parse depth layers	Dec 6, 2016
webcam_demo.lua	webcam_demo.lua	Pass bool to setup_gpu at other call sites	Oct 13, 2016

Provide feedback