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.
VRM4U is a comprehensive Unreal Engine plugin (UE4/UE5) for importing, rigging, and animating VRM (Virtual Reality Model) avatars. It provides both editor-time import workflows and runtime components for real-time character rendering and animation. The plugin supports VRM 0.x and VRM 1.0 specifications, along with related formats like GLTF/GLB, BVH, PMX, and VRMA. Core features include: skeletal mesh generation with humanoid bone mapping, MToon/PBR material conversion, morph target (blend shape) import with ARKit and MetaHuman mappings, spring bone physics simulation, ControlRig generation, and motion capture integration (VMC protocol, Sony mocopi, Oculus hand tracking). The codebase is organized into multiple UE modules (VRM4U, VRM4ULoader, VRM4UImporter, VRM4URender, VRM4UCapture) and includes both C++ anim nodes and Blueprint/Python automation scripts. The architecture follows a standard UE plugin pattern with editor and runtime separation. The import pipeline (VRM4UImporterFactory + VRMConverter) uses the bundled Assimp library for model parsing and RapidJSON for VRM metadata extraction, then procedurally creates UE assets (skeletal meshes, physics assets, animation blueprints, control rigs). The runtime module provides custom animation nodes for spring bones, constraints, VMC, and retargeting, along with a scene view extension for rim-light post-processing and GBuffer capture. The plugin also exposes a Blueprint function library for high-level avatar control (morph targets, material overrides, camera followers). Despite its depth, the codebase shows signs of maintenance burden: extensive #if UE_VERSION_OLDER_THAN guards for supporting UE 4.20 through 5.8, reliance on private engine rendering headers, and monolithic conversion functions. The bundled third-party dependencies (Assimp, RapidJSON) are fixed versions that require manual updates. However, the plugin is actively developed with frequent commits and broad feature coverage, making it a valuable resource for UE developers integrating VRM avatars.
This repository provides a comprehensive research codebase for AI-driven character animation, spanning multiple SIGGRAPH publications from 2017 to 2024. It includes deep learning model training scripts (TensorFlow for PFNN, MANN, NSM; PyTorch for GNN, PAE, Codebook Matching) and corresponding Unity integration for real-time inference and visualization. The codebase ingests motion capture data (BVH, FBX, TRC), processes it through a Unity-based data pipeline (MotionData, Frame, Modules), trains phase-functioned or mode-adaptive networks, and exports trained weights for runtime use. The repository is hybrid: primarily a research project with both data pipeline (motion capture processing, feature extraction) and application (Unity demos, IK, visualization tools) components. It is not a standalone library but a reference implementation for academic publications, making it valuable for researchers and developers working on character animation, deep learning for motion, and real-time animation systems.