sen-mao / StyleDiffusion

Official Implementations "StyleDiffusion: Prompt-Embedding Inversion for Text-Based Editing" (CVMJ2024)
55 stars 1 forks source link

Why did my style transfer fail? #6

Closed ScuLancer closed 2 months ago

ScuLancer commented 3 months ago

hello,I have the question about the code This is my Inferential code python stylediffusion.py --is_train '' --index 1 --prompt "a tent in forest" \ --image_path "./example_images/a tent in forest.jpg" \ --target "a tiger in forest" \ --tau_v [.2,] --tau_c [.6,] --tau_s [.6,] --tau_u [.5,] \ --blend_word "[('tent',), ('tiger',)]" --eq_params "[('tiger',), (2,)]" --edit_type Replacement image But you can see that I failed the style Transfer. I want to know why.Thank you very much!

sen-mao commented 3 months ago

Hi, you can try the code in p2plus catalogue:

python text_inv_w_p2plus.py --image_path "./example_images/a tent in forest.jpg" \
                            --prompt "a tent in forest" --target "a tiger in forest" \
                            --tau_c "[.6,]"  --tau_s "[.6,]"  --tau_u "[.0,.1,.2,.3,.4,.5,.6,.7,.8,.9,1.,]" \
                            --blend_word "[('tent',), ('tiger',)]"  --eq_params "[('tiger',), (2,)]" \
                            --edit_type Replacement