This repository implements VPoser, a variational autoencoder-based human pose prior for SMPL/X body models. It provides tools to encode/decode human poses into a low-dimensional latent space, sample novel poses, and perform inverse kinematics fitting to 2D/3D key points using the learned prior. The codebase includes data preparation scripts for the AMASS dataset, a PyTorch Lightning trainer, a flexible IK engine with LBFGS/Adam optimizers, and utilities for rotation mathematics and linear blend skinning. The system is designed as a reusable library for pose prior applications in research and graphics. Architecture-wise, the project is organized into modules: body_model (SMPL/X loading and LBS), models (VPoser VAE and IK engine), tools (rotation conversions, 6D representation, file/path helpers), data (dataset loading and preparation), train (training scripts), and tutorials (Jupyter notebooks and example scripts). Entry points include training scripts (e.g., train/V02_05/V02_05.py), IK demonstrations (tutorials/ik_example_*.py), and usage tutorials (tutorials/vposer.ipynb). External dependencies are primarily PyTorch, Lightning, OmegaConf, and optional visualization packages. Data flows: For training, AMASS .npz files are converted to .pt tensors via prepare_data.py, loaded by VPoserDS, and fed into VPoserTrainer. For inference, pretrained models are loaded via model_loader.py, enabling pose encoding/decoding or IK fitting. The IK engine iteratively optimizes pose, shape, and translation in the VPoser latent space to match target key points.
SkinTokens is a research codebase for a novel 3D character rigging method (TokenRig) that learns a compact discrete representation (SkinTokens) for skinning weights. The system ingests 3D meshes with skeleton and skinning data (from ArticulationXL, VRoid Hub, ModelsResource), transforms them through a pipeline of augmentation, order arrangement, vertex group extraction, and sampling, then trains an FSQ-CVAE to compress skinning weights into discrete tokens. A Qwen3-0.6B-based autoregressive transformer (TokenRig) generates the full rig (skeleton hierarchy + skin tokens) as a single token sequence. The repo includes a Blender-based server for mesh loading and export, a Gradio demo, a CLI, and extensive data processing modules. The code is well-structured with dataclass-based configs, abstract interfaces, and modular design, but remains primarily a research artifact with some incomplete training implementations (e.g., TokenRig training_step raises NotImplementedError). Its hybrid nature leans strongly toward data research, as the core contribution is a learned representation and generative model for animation rigging.
This repository is a curated 'awesome list' (bibliography) of research papers, datasets, and software tools at the intersection of artificial intelligence and molecular dynamics/conformational analysis. It contains no executable code, data files, or notebooks—only a single README.md that organizes hundreds of references by topic (e.g., neural network potentials, force fields, generative models for protein conformations, AlphaFold-based methods). The repo serves as a comprehensive literature index for researchers in computational chemistry and structural biology, but it is not a functional software project or data pipeline. No raw data, transformation scripts, or analytical models are present; reproducibility is not applicable.