# iamcreateai/neoverse architecture

> Architecture diagram of the GitHub repository iamcreateai/neoverse, generated by GitDiagram from its file tree, README and a sample of source files (last updated 2026-09-29). It is an AI-made overview: check details against the source.

- Interactive diagram: https://gitdiagram.com/iamcreateai/neoverse
- Repository: https://github.com/iamcreateai/neoverse

## Overview

NeoVerse reconstructs a scene from an in-the-wild image or video, lets users design a camera trajectory, renders target-view guidance, and generates a novel-trajectory video. Its runtime combines a 3D reconstructor, camera/geometry utilities, and a Wan video-generation pipeline; the README also documents a Gradio workflow, plug-in reconstructors, and ControlBranch fine-tuning. The excerpts show the CLI’s reconstruction and generation calls and the app’s trajectory-preview flow; pipeline internals and some UI wiring were not sampled, so those links remain at subsystem level.

Main entry points: [`diffsynth/pipelines`](https://github.com/iamcreateai/neoverse/tree/main/diffsynth/pipelines) (NeoVerse pipeline), [`inference.py`](https://github.com/iamcreateai/neoverse/blob/main/inference.py) (CLI inference), [`app.py`](https://github.com/iamcreateai/neoverse/blob/main/app.py) (Gradio demo), [`diffsynth/auxiliary_models/worldmirror/models/models/worldmirror.py`](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/auxiliary_models/worldmirror/models/models/worldmirror.py) (WorldMirror reconstructor)

## Diagram (Mermaid)

```mermaid
flowchart TD

subgraph group_workflow["User Workflow"]
  node_cli["CLI inference<br/>[inference.py]"]
  node_demo["Gradio demo<br/>[app.py]"]
  node_trajectory["Trajectory tools<br/>[auxiliary.py]"]
  node_preview["Scene preview<br/>[app.py]"]
  node_output["Generated video"]
end

subgraph group_reconstruct["Scene Reconstruction"]
  node_reconstructor["WorldMirror reconstructor<br/>[worldmirror.py]"]
  node_depth["Depth Anything 3<br/>[api.py]"]
  node_scene["Gaussian scene data<br/>[app.py]"]
end

subgraph group_generation["Video Generation"]
  node_pipeline["NeoVerse pipeline"]
  node_wan["Wan video model<br/>[wan_video_dit.py]"]
  node_control["ControlBranch"]
  node_camera["Camera controller"]
  node_motion["Motion controller"]
  node_modelmanager["Model manager<br/>[model_manager.py]"]
end

subgraph group_training["Model Training"]
  node_trainentry["Training entry<br/>[train.py]"]
  node_trainer["Text-image trainer<br/>[text_to_image.py]"]
  node_datasets["Training datasets"]
end

node_user(("Image/video user"))

node_user -->|"provides input"| node_cli
node_user -->|"uploads media"| node_demo
node_cli -->|"builds trajectory"| node_trajectory
node_demo -->|"designs trajectory"| node_trajectory
node_cli -->|"invokes generation"| node_pipeline
node_demo -->|"renders preview"| node_preview
node_cli -->|"reconstructs views"| node_reconstructor
node_reconstructor -->|"returns splats"| node_scene
node_preview -->|"visualizes splats"| node_scene
node_pipeline -->|"returns video"| node_output
node_modelmanager -->|"loads models"| node_pipeline
node_trainentry -->|"starts training"| node_trainer
node_trainer -->|"uses data"| node_datasets
node_depth -.->|"optional backend"| node_reconstructor
node_pipeline -.->|"uses control branch"| node_control
node_pipeline -.->|"uses camera control"| node_camera
node_pipeline -.->|"uses motion control"| node_motion

click node_cli "https://github.com/iamcreateai/neoverse/blob/main/inference.py"
click node_demo "https://github.com/iamcreateai/neoverse/blob/main/app.py"
click node_trajectory "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/utils/auxiliary.py"
click node_preview "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/utils/app.py"
click node_pipeline "https://github.com/iamcreateai/neoverse/tree/main/diffsynth/pipelines"
click node_wan "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_dit.py"
click node_control "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_neoverse_controller.py"
click node_camera "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_camera_controller.py"
click node_motion "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_motion_controller.py"
click node_reconstructor "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/auxiliary_models/worldmirror/models/models/worldmirror.py"
click node_depth "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/auxiliary_models/depth_anything_3/api.py"
click node_scene "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/utils/app.py"
click node_modelmanager "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/model_manager.py"
click node_trainentry "https://github.com/iamcreateai/neoverse/blob/main/train.py"
click node_trainer "https://github.com/iamcreateai/neoverse/blob/main/diffsynth/trainers/text_to_image.py"
click node_datasets "https://github.com/iamcreateai/neoverse/tree/main/training/data"

classDef toneNeutral fill:#f8fafc,stroke:#334155,stroke-width:1.5px,color:#0f172a
classDef toneBlue fill:#dbeafe,stroke:#2563eb,stroke-width:1.5px,color:#172554
classDef toneAmber fill:#fef3c7,stroke:#d97706,stroke-width:1.5px,color:#78350f
classDef toneMint fill:#dcfce7,stroke:#16a34a,stroke-width:1.5px,color:#14532d
classDef toneRose fill:#ffe4e6,stroke:#e11d48,stroke-width:1.5px,color:#881337
classDef toneIndigo fill:#e0e7ff,stroke:#4f46e5,stroke-width:1.5px,color:#312e81
classDef toneTeal fill:#ccfbf1,stroke:#0f766e,stroke-width:1.5px,color:#134e4a
class node_cli,node_demo,node_trajectory,node_preview,node_output,node_user toneBlue
class node_reconstructor,node_depth,node_scene toneAmber
class node_pipeline,node_wan,node_control,node_camera,node_motion,node_modelmanager toneMint
class node_trainentry,node_trainer,node_datasets toneRose
```

## Components

### User Workflow

- [CLI inference](https://github.com/iamcreateai/neoverse/blob/main/inference.py): `inference.py`
- [Gradio demo](https://github.com/iamcreateai/neoverse/blob/main/app.py): `app.py`
- [Trajectory tools](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/utils/auxiliary.py): `diffsynth/utils/auxiliary.py`
- [Scene preview](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/utils/app.py): `diffsynth/utils/app.py`
- **Generated video**

### Scene Reconstruction

- [WorldMirror reconstructor](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/auxiliary_models/worldmirror/models/models/worldmirror.py): `diffsynth/auxiliary_models/worldmirror/models/models/worldmirror.py`
- [Depth Anything 3](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/auxiliary_models/depth_anything_3/api.py): `diffsynth/auxiliary_models/depth_anything_3/api.py`
- [Gaussian scene data](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/utils/app.py): `diffsynth/utils/app.py`

### Video Generation

- [NeoVerse pipeline](https://github.com/iamcreateai/neoverse/tree/main/diffsynth/pipelines): `diffsynth/pipelines`
- [Wan video model](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_dit.py): `diffsynth/models/wan_video_dit.py`
- [ControlBranch](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_neoverse_controller.py): `diffsynth/models/wan_video_neoverse_controller.py`
- [Camera controller](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_camera_controller.py): `diffsynth/models/wan_video_camera_controller.py`
- [Motion controller](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/wan_video_motion_controller.py): `diffsynth/models/wan_video_motion_controller.py`
- [Model manager](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/models/model_manager.py): `diffsynth/models/model_manager.py`

### Model Training

- [Training entry](https://github.com/iamcreateai/neoverse/blob/main/train.py): `train.py`
- [Text-image trainer](https://github.com/iamcreateai/neoverse/blob/main/diffsynth/trainers/text_to_image.py): `diffsynth/trainers/text_to_image.py`
- [Training datasets](https://github.com/iamcreateai/neoverse/tree/main/training/data): `training/data`

### Other components

- **Image/video user**

## Connections

- Image/video user → CLI inference: provides input
- Image/video user → Gradio demo: uploads media
- CLI inference → Trajectory tools: builds trajectory
- Gradio demo → Trajectory tools: designs trajectory
- CLI inference → NeoVerse pipeline: invokes generation
- Gradio demo → Scene preview: renders preview
- CLI inference → WorldMirror reconstructor: reconstructs views
- WorldMirror reconstructor → Gaussian scene data: returns splats
- Scene preview → Gaussian scene data: visualizes splats
- NeoVerse pipeline → Generated video: returns video
- Model manager → NeoVerse pipeline: loads models
- Training entry → Text-image trainer: starts training
- Text-image trainer → Training datasets: uses data
- Depth Anything 3 → WorldMirror reconstructor: optional backend
- NeoVerse pipeline → ControlBranch: uses control branch
- NeoVerse pipeline → Camera controller: uses camera control
- NeoVerse pipeline → Motion controller: uses motion control