xcyan / eccv18_mtvae Public

Tensorflow Implementation of ECCV'18 paper: Multimodal Human Motion Synthesis

github.com/xcyan/eccv18_mtvae

Notifications

Name		Name	Last commit message	Last commit date
Latest commit History 12 Commits
nets		nets
preprocess		preprocess
H36M_BaseAnalogyModel.py		H36M_BaseAnalogyModel.py
H36M_BasePredModel.py		H36M_BasePredModel.py
H36M_MTVAEAnalogyModel.py		H36M_MTVAEAnalogyModel.py
H36M_MTVAEPredModel.py		H36M_MTVAEPredModel.py
H36M_VAEPredModel.py		H36M_VAEPredModel.py
H36M_modelfile.py		H36M_modelfile.py
H36M_testfile.py		H36M_testfile.py
LICENSE		LICENSE
README.md		README.md
h36m_analogy.py		h36m_analogy.py
h36m_gensample.py		h36m_gensample.py
h36m_input.py		h36m_input.py
h36m_losses.py		h36m_losses.py
h36m_train_mtvae.py		h36m_train_mtvae.py
h36m_train_vae.py		h36m_train_vae.py
init.sh		init.sh
model_utils.py		model_utils.py
utils.py		utils.py

Repository files navigation

MT-VAE for human motion synthesis

This is the code for ECCV 2018 paper MT-VAE: Learning Motion Transformations to Generate Multimodal Human Dynamics by Xinchen Yan, Akash Rastogi, Ruben Villegas, Kalyan Sunkavalli, Eli Shechtman, Sunil Hadap, Ersin Yumer, Honglak Lee.