Plugins + Scripts AI-Driven Skeleton Engine (BoneReceiver)

This is an experimental project to drive Virt-A-Mate (VaM) skeleton animations using a Python-based CVAE Machine Learning engine.

[ Quick Start Guide ]
1. Download the ML Skeleton Engine (Python)
[ https://github.com/yuyaslabo/VaM_ML_Skeleton_Engine ] (Open "Releases")
2. Unzip the file and run START_ENGINE.bat. ( Recommended to place the unzipped folder in the root directory of your drive. Example: c:\VaM_Skeleton_Engine_v1.0 )
3. In VaM, add AIVRLab.BoneReceiver.cs to your Person atom.

That's it! Your character will start moving.

[ Note ]
- Optimized for Person Scale (height) about 0.9 (adjust morphs/height). If used at the default Scale 1.0, the waist may appear bent.
- Please make sure that the "Allow Plugins Network Access" option is checked. (In VaM “User Preferences” > “Security”)
- Port: Default 9998 (Changeable in Plugin GUI and target_ip.csv).
- Offloading: Supports running the Engine on a separate LAN PC. (Need edit target_ip.csv)

[ Credits & Acknowledgments ]
This model was trained using the following community assets. I am deeply grateful to these creators for their contributions to the VaM ecosystem:
- lapiro: Voxta Demo
- Acid Bubbles: Voxta Beta Scene
- PrimeMocap: Girlfriend Dance, Halloween Free,
- NekoNeko: Riding Challenge(Hard) - Vam Launch
- LDR: Evil Vixen V2, Wet Whispers, Game for lovers 2
- NLvam: NL077 Girl tossing and turning
- VamTimbo: Elf Breast Massage, Focus On Me K-Pop
- CuddleMocap: Cute Idles Simplified
[ Demo Video Assets ]
- Xspada: Sugar (Character Look used in the video)
The training data consists exclusively of Free/CC-licensed content; no Paid/Premium assets were used. To respect the original licenses, this resource is distributed under CC BY-NC-ND.

[ Setup & Demo Video ]

Note: I am Japanese, so I used AI to assist in creating this description.