The Code will be released soon!
This repo is the official implementation of ConvFormer: 3D Human Pose Estimation with Dynamic Multi-headed Convolutional Attention
The code was developed and tested under the following environment:
- Python 3.8.5
- Pytorch 1.10.0
- CUDA 10.2.0
You can create the environment running:
conda env create -f convformer.yml
Our code is compatible with the dataset setup introduced in Martinez et al. and Pavllo et al. Please refer to either repo to set up H3.6M (./data director)
