SHIVA

1. Skillset Hub with Intelligence Virtual Academy

2. Scalable Hybrid Intelligence in a Virtual Agency

3. Sharing Hub and Incubator with Virtual Agents

Folder Structure

SHIVA

📦 SHIVA
├─ 📂 apps       # Submodules / local clones of AI apps
│   ├─ 📂 <service_group>
│   │   ├─ 📂 <service_name>
│   │   └─ 📂 …
│   └─ 📂 …
├─ 📂 gateway    # Master API gateway
│   ├─ 📄 config.py       # Service loading logic
│   ├─ 📄 main.py         # Loads routers from `apps`
│   └─ 📄 utils.py        # Dynamic import helpers
├─ 📂 uinified   # Unified Gradio interface
│   ├─ 📄 config.py       # UI metadata config
│   ├─ 📄 main.py         # Builds dynamic UI from `apps`
│   └─ 📄 utils.py        # Dynamic import helpers
├─ 📂 configs    # Service discovery config
│   ├─ 📄 apps.yaml       # [Optional] app config
│   └─ 📄 services.yaml   # Service Registry / Discovery
├─ 📂 controller
│   ├─ 📄 orchestrator.py
│   └─ 📄 utils.py
├─ 📄 docker-compose.yaml
└─ …

AI Service

📦 <service_name>
├─ 📂 src        # Core AI Logic - reuse original source code
├─ 📂 fapi        # FastAPI Application
| ├─ 📄 api.py      # Inference routes
| ├─ 📄 config.py   # Configs like model paths, constants
| ├─ 📄 main.py     # FastAPI entrypoint
| └─ 📄 service.py  # Core logic for model dispatch + inference
├─ 📂 grui      # Gradio app for user interface
| ├─ 📂 assets      # Images, stylesheets
| ├─ 📄 main.py     # Gradio interface layout
| └─ 📄 utils.py    # Helpers for UI rendering
├─ …
├─ 📄 Dockerfile
├─ 📄 README.md         # Documentation
└─ 📄 requirements.txt  # Libraries to install

Services

Composition

• AI Agent (GenArtist)

Language

• Multi-modal Dialogue

• Role Play (RealChar | LLM-Roleplay)

• Settlement Simulation (Neighborly)

• Story Generation (MoPS)

Toolkit

• Background Decomposition (RemBg)

• Instance Segmentation (OWL-ViT+SAM | Florence+SAM)

• 3D Object Segmentation (Point-SAM)

• Video Matting (RobustVideoMatting | MatAnyone)

• Controlnet Preprocessors (ControlNet)

• OpenPose Editor (2D Editor | 3D Editor)

• Image Upsampling (Real-ESRGAN | GFPGAN)

Vision

• Image Generation

• Image Editing (MagicQuill)

• Re-Lighting (ICLight)

• Style Transfer (CSGO)

• Image Restoration (DeOldify | SUPIR)

• Face Editor (SD-Face-Editor)

Apps

• Live Painting (Krita.ai)

Multi-View Synthesis

• Text-to-MV (MVDream)

• Image-to-MV (Zero123++)

• Image/Text-to-Panorama (SD-T2I-360PanoImage | MVDiffusion | 360DVD)

• Camera Control (SeVa | CameraCtrl | MotionCtrl)

• 3D Visualization (Viser | Camera Trajectory | HandyTrajDrawer)

Notation

• MV: Multi-view

3D Reconstruction

• Image-to-3D-Anime-Character (CharacterGen)

• Image-to-3D (Unique3D | StableFast3D | InstantMesh)

• Multi-Instance-to-3D (MIDI-3D)

Animation

• 2D Animation (Animate Anything | MOFA-Video)

• 3D Animation (Make It Animatable)

• Face Animation (X-Pose | MimicTalk | GeneFace++ | Real3DPortrait)

• Human Animation (StableAnimator | Moore-AnimateAnyone)

• Lip Sync (Hallo)

Human Motion

• Wrapper

• Human Motion (MDM | PriorMDM | SinMDM | MoMo)

• Human-Object Interaction (Text2HOI | HOI-Diff)

Virtual Try-on (ViTon)

• Multi-View ViTon (MV-VTON)

• Full-Body ViTon (OOTDiffusion | PICTURE)

Sound

• Audio Generation (Make-An-Audio | AudioLDM | Amphion | AudioCraft | Suno|Bark)

• Music Generation (MusicGPT | YuE | Riffusion)

• Voice Clone (Fish Speech | OpenVoice | Real-Time-Voice-Cloning | Bark-Voice-Clone)

• Voice Clone VN (viXTTS | EraX-Female)

Coding

• Wireframe-to-Website (Pix2Code | Screenshot2Code by Emil Wallner | Screenshot2Code by Abi)

Misc.

• Raster-to-Vector (VTracer | Im2Vec)

• Font Generation (DeepVecFont-v2 | DeepVecFont)

References

• WebUI (WebUI-A1111 | Dialog-WebUI)

• Digital Clone (HeyGem)