MagicSource

@maowayne123 thank u. But your repo is based on certain commit isn't it? does this repo's mmpose able to reproduce your papre result？

@LegendBC Hi，NuScenes有这个GT标注吗？那是不是可以把这个boundary和detection联合起来一起训

@liuruijin17 is there any plan on when will code public?

@kmonachopoulos @firasomran01 Deeplearning is just over fit everything........... out of dataset can not works well

SO, as a stupid user. how can I run `nvim` after I just runed `winget install neovim` **extacly** same as any unix system??

@LizhenWangT then does faceverse able to generates 51 PCA same as Apple's and align with apples expression deifinition? (so that many models based on apple's definition can be driven)

@LizhenWangT that is promising! Are u trying using Apple's standared face model to fit FaceVerse's version? It seems once we have a dividable 52 shapes, and it's semantic meaning, we...

@LizhenWangT I am hoping for your impressive result!

Hi, the onnx exported from yolov7-d2, should already included prerpocess, which means, you should not need permute HWC to CHW, and not need to do normalization. it was inside model...

Please try narrow down the lr.