HAnim, glTF, and VRM—three major frameworks for representing and animating humanoid characters. This appendix is designed to accompany your white paper and can be integrated into Web3D Consortium documentation, conference submissions, or standards outreach.

It is structured, deep, and aligned with your communication style: modular, technical, and strategic.

A. Overview of the Three Systems

• HAnim → semantic skeleton + motion interoperability
• glTF → efficient asset packaging + runtime delivery
• VRM → avatar identity + expressive metadata

Together, they form a complementary ecosystem.

B. Structural Comparison

B.1 Skeleton Hierarchy

HAnim

• Normative joint hierarchy
• Standardized joint names
• Anatomically grounded (ISO 7250, CAESAR)
• Includes Joint, Segment, Site nodes
• Default pose is normative

glTF

• Skeletons defined via nodes and skins
• Joint names not standardized
• No anatomical semantics
• Default pose varies by asset
• Skinning via linear blend skinning

VRM

• Based on glTF skeleton structure
• Adds VRM Humanoid mapping:
  • Maps arbitrary joint names → standardized humanoid bones
• Requires humanoid pose compliance
• Includes constraints for avatar consistency
• Key Insight: HAnim provides semantic rigor; glTF provides delivery efficiency; VRM provides avatar consistency.

 

B.2 Feature Points / Sites

HAnim

• Rich set of HAnimSite nodes
• Includes anatomical landmarks:
  • Eyes, ears, nose, chin
  • Hand/wrist attachment points
  • Foot/ankle markers
• Used for:
  • Sensors
  • Attachments
  • Motion capture alignment

glTF

• No concept of feature points
• Attachments require custom nodes

VRM

• Provides VRM BlendShape and VRM SpringBone
• Facial landmarks handled via blendshape presets
• No anatomical feature point standard

Key Insight: HAnim is the only system with normative anatomical feature points.

C. Motion and Animation Semantics

C.1 Motion Data

HAnim

• ISO/IEC 19774‑2 defines motion data formats
• Supports:
  • Keyframe interpolation
  • Frame‑based motion streams
  • Declarative animation via X3D event model
• Designed for interoperable motion reuse

glTF

• Supports animation channels:
  • Translation
  • Rotation
  • Scale
• No semantic constraints
• Motion retargeting requires external tools

VRM

• Inherits glTF animation
• Adds:
  • Facial expression presets
  • SpringBone physics
• Motion semantics not standardized

Key Insight: HAnim is the only system with standardized motion semantics.

D. Avatar Identity and Metadata

HAnim

• Focuses on anatomy and motion
• No avatar identity layer
• Metadata handled via X3D extensions

glTF

• Minimal metadata
• Extensions required for identity or semantics

VRM

• Rich avatar identity model:
  • Name, author, license
  • Facial expressions
  • Hair/cloth physics
  • First‑person view configuration
  • Look‑at behavior
• Designed for social VR and metaverse platforms

Key Insight: VRM provides the identity layer missing from HAnim and glTF.

E. Encoding and Runtime Behavior

F. Interoperability Matrix

F.1 HAnim ↔ glTF

• glTF can carry HAnim skeletons if:
  • Joint names follow HAnim conventions
  • Default pose is preserved
• HAnim motion can be retargeted to glTF skeletons
• glTF lacks feature points → must be added manually

F.2 HAnim ↔ VRM

• VRM Humanoid mapping can map HAnim joints
• VRM facial expressions can complement HAnim skeletons
• HAnim feature points can enhance VRM attachment logic

F.3 glTF ↔ VRM

• VRM is built on glTF
• glTF → VRM conversion requires:
  • Humanoid bone mapping
  • Metadata injection
  • Blendshape presets

G. Strengths and Weaknesses Summary

H. Strategic Alignment for Web3D / Cultural Heritage / Open Metaverse

HAnim

• Best for interoperable human animation, training, simulation, cultural heritage reenactments
• Provides semantic rigor needed for scientific and educational applications

glTF

• Best for asset delivery, rendering pipelines, and lightweight web deployment
• Ideal for distributing HAnim‑based models

VRM

• Best for avatars, identity, social VR, metaverse platforms
• Complements HAnim by adding expressive layers

I. Recommended Integration Strategy

1. Use HAnim for:

• Skeleton definition
• Motion semantics
• Anatomical accuracy
• Feature points

2. Use glTF for:

• Packaging HAnim models
• Efficient transmission
• Runtime rendering

3. Use VRM for:

• Avatar identity
• Facial expressions
• Social VR compatibility

Combined Stack Example: HAnim skeleton + glTF asset + VRM identity → fully interoperable avatar pipeline.

J. Example Crosswalk Table (HAnim → glTF → VRM)

This mapping enables automated conversion pipelines.

K. Conclusion

HAnim, glTF, and VRM each solve different parts of the humanoid representation problem:

• HAnim → semantic skeleton + motion interoperability
• glTF → efficient asset delivery
• VRM → avatar identity + expressive metadata

Together, they form a powerful, complementary ecosystem for Web3D, cultural heritage, simulation, and the open metaverse.