VIBE
VIBE copied to clipboard
Official implementation of CVPR2020 paper "VIBE: Video Inference for Human Body Pose and Shape Estimation"
I tried maskrcnn and yolov3 with default multi-person-tracker https://github.com/mkocabas/multi-person-tracker I think it's not good on some video (for example, people that appear small in the basketball video). Are there any...
Hi I am aware that the pose parameters that you give in the smpl are present in terms of angles. I want to know in what type of axis are...
I am replicating the Vid2Dopler project. VIBE is the first step in the pipeline. However, the version of the related dependency package is too low to run on NVIDIA GeForce...
Operating system: Ubuntu 22.04 LTS Python 3.7 (the environment provided was installed, Torch 1.4) Running the demo caused a CUDA runtime error. "no kernel image is available for execution on...