Skip to content
ProGamerGov edited this page Feb 18, 2017 · 4 revisions

Images used:

fig4_content.jpg: https://github.com/leongatys/NeuralImageSynthesis/blob/master/Images/ControlPaper/fig4_content.jpg

Fine style: https://github.com/leongatys/NeuralImageSynthesis/blob/master/Images/ControlPaper/fig4_style2.jpg

Course style: https://github.com/leongatys/NeuralImageSynthesis/blob/master/Images/ControlPaper/fig4_style3.jpg


Low Res Stylemix Image:

th neural_style.lua -tv_weight 0 -image_size 1296 -content_weight 0E+00 -style_weight 1.0E+05 -output_image out5_hr.png -num_iterations 550 -content_image result_2.png -style_image result.png -content_layers relu2_1,relu4_2 -style_layers relu2_1,relu4_2 -save_iter 0 -print_iter 50 -seed 876 -init image -backend cudnn -cudnn_autotune

High Res Stylemix Image:

th neural_style.lua -init_image out5_hr.png -tv_weight 0 -image_size 1296 -content_weight 0E+00 -style_weight 1.0E+05 -output_image out5_hr_2100.png -num_iterations 550 -content_image result_2.png -style_image result.png -content_layers relu2_1,relu4_2 -style_layers relu2_1,relu4_2 -save_iter 0 -print_iter 50 -seed 876 -init image -backend cudnn -cudnn_autotune

Low Res Output Image:

th neural_style.lua -style_weight 2.0E+08 -content_weight 1.0E+05 -output_image out_final.png -num_iterations 1000 -content_image fig4_content.jpg -style_image out5_hr.png -image_size 512 -save_iter 0 -print_iter 50 -seed 876 -init image -backend cudnn -cudnn_autotune

High Res Output Image:

th neural_style.lua -style_weight 2.0E+08 -content_weight 1.0E+05 -output_image out_final_hr.png -num_iterations 550 -content_image fig4_content.jpg -init_image out_final.png -style_image out5_hr_2100.png -image_size 1536 -save_iter 0 -print_iter 50 -seed 876 -init image -backend cudnn -cudnn_autotune

Clone this wiki locally