Gengzigang / PCT

This is an official implementation of our CVPR 2023 paper "Human Pose as Compositional Tokens" (https://arxiv.org/pdf/2303.11638.pdf)
MIT License
332 stars 21 forks source link

希望得到建议 #2

Closed EiJuN-xmu closed 1 year ago

EiJuN-xmu commented 1 year ago

现在我训练好了模型,并成功运行了dist_test.sh。我应该怎么做,来实现demo中的效果呢?即如何输出关键点提取后的图像? 希望能得到一些建议