Open Nortnb opened 11 months ago
hi @cuiaiyu, sorry for bothering, could you give me some insights ? i have added my picture, keypoints, and human parse into dataset
and then i run a try-on:
%matplotlib inline
pid = ("fashionMENDenimid0000387902_7additional.jpg", 4, None) # load the 3-rd person from "pattern" group, NONE (no) garment is interested
gids = [
("fashionWOMENDressesid0000111101_4full.jpg",0,3), # load the 0-th person from "plaid" group, garment #5 (top) is interested
("plain",1,1), # load the 3-rd person from "pattern" group, garment #1 (bottom) is interested
]
# tuck in (dressing order: hair, top, bottom)
pimg, gimgs, oimgs, gen_img, pose = dress_in_order(model, pid, gids=gids, order=[2,5,1])
plot_img(pimg, gimgs, gen_img=gen_img, pose=pose)
but my garment is still unrecognized:
by the way this is the garment i'm testing with:
thanks
can you help me with,how to run code and give input i am new here
Hi, thank you for the great work,
I have played around with the colab, and the results are great so far, so now i want to test it using my own images, i've read this issue about custom image, and there it's mentioned that i need to "first get pose estimation and human parse, so you have (img, parse, pose) pairs for the target person and each target garment."
currently i have only the clothe images,
i'm sorry if this is too basic, the tools are probably in the repo already, but i can't figure out how to use them, thank you so much.