search

cure-lab / DeciWatch

[ECCV 2022] Official implementation of the paper "DeciWatch: A Simple Baseline for 10x Efficient 2D and 3D Pose Estimation"
Apache License 2.0
177 stars 15 forks source link

How to do a single denoising prediction for a given set of joint (x,y) values? #10

Closed theneuronprogrammer closed 2 years ago

theneuronprogrammer commented 2 years ago

i/p - [(x1, y1), (x2, y2), ...... , (xn, yn)]

I want output as -

o/p - [(x1_corrected, y1_corrected), (x2_corrected, y2_corrected), ...... , (xn_corrected, yn_corrected)]

juxuan27 commented 2 years ago

Hi, @theneuronprogrammer ! Thank you for your focus! Do you mean how to denoise the predicted poses without sampling? If so, simply changing the sample interval to 1 would work. We have also provided corresponding checkpoints on google drive and baidu disk.