search

twehrbein / Probabilistic-Monocular-3D-Human-Pose-Estimation-with-Normalizing-Flows

This is the official implementation of the ICCV 2021 Paper "Probabilistic Monocular 3D Human Pose Estimation with Normalizing Flows" by Tom Wehrbein, Marco Rudolph, Bodo Rosenhahn and Bastian Wandt.
MIT License
70 stars 6 forks source link

Excluding Some Human3.6M CDF Data #5

Closed GloryyrolG closed 1 year ago

GloryyrolG commented 1 year ago

Hi @twehrbein ,

Found you did really good work! May I ask the reason for excluding some subsets of Human3.6M,

https://github.com/twehrbein/Probabilistic-Monocular-3D-Human-Pose-Estimation-with-Normalizing-Flows/blob/ad2fdf21da1dfb689a5c2a531ebda6e23d95ccc0/create_data.py#L172

Thanks & regards,

twehrbein commented 1 year ago

Hey, thanks for your interest!

To be honest I don't remember why I excluded the sequences. Maybe they somehow got corrupted during preprocessing/downloading.

Since it's only a really small percentage of the whole dataset, it shouldn't matter performance wise.