search

cuiaiyu / dressing-in-order

(ICCV'21) Official code of "Dressing in Order: Recurrent Person Image Generation for Pose Transfer, Virtual Try-on and Outfit Editing" by Aiyu Cui, Daniel McKee and Svetlana Lazebnik
https://cuiaiyu.github.io/dressing-in-order
Other
507 stars 126 forks source link

Human Parser Issue #59

Closed vinodbukya6 closed 1 year ago

vinodbukya6 commented 1 year ago

Hi @cuiaiyu I hope you are doing well. Can you please help me regarding human parser methods. I have created parser using SCHP(On CPU not working), so using LIP JPPNet Parser(https://github.com/Engineering-Course/LIP_JPPNet). Exatracted Parser using SCHP colab notebook and results are Great. Not able to generate same results with LIP JPPNet Parser.

Is your demo works with any method of Human Parser or only works with SCHP parser(which only works on GPU)?

Original Image: test_men04

SCHP Colab Notebook Result: test_men04

LPP JPPNet Parser Result: test_men04

cuiaiyu commented 1 year ago

As long as you have LIP labels (with each category in the same order), you should be good.

If you are different label than LIP, you can change the index of garments and the part of interested in this file. https://github.com/cuiaiyu/dressing-in-order/blob/main/datasets/human_parse_labels.py

vinodbukya6 commented 1 year ago

Hi, everything is working fine. Implemented Human Parser using this GitHub Repo: https://github.com/samuelesabella/Self-Correction-Human-Parsing

Thank you so much.