This guide explains how to use "Hoshizuna: Model Viewer for Dome", a 3D viewer for projecting 3D models in planetariums and dome theaters. While basic usage is described in the application's built-in help screen, this manual provides more specific operational procedures and configuration methods along with screenshots. Please note that since this manual is being written in parallel with development, it may contain screenshots from different versions or feature unreleased functions.
Version at the time of article update: 1.72
Language: Japanese/ English
When it comes to projecting 3D models into a dome space like a planetarium and moving them in real-time, you might think that special computers or software are required. However, nowadays, anyone can easily do this as long as they have a laptop with a web browser. Hoshizuna is an application created as a single HTML file. By preparing a 3D model from platforms like Sketchfab and accessing Hoshizuna, an immersive full-dome space begins.
Hoshizuna is developed with the aim of being easy for anyone to operate. However, as features have been expanded, it has become impossible to describe all operational procedures in the built-in help. Therefore, we decided to create a more detailed manual with screenshots. This manual is structured to correspond to the application's screen layout and features. You can read it in order from the beginning, or search for features in the table of contents and refer only to what you need. Also, as it is still under development, we intend to further enrich the content of this page in the future.
Hoshizuna: Model Viewer for Dome
https://tasukuyoshioka.github.io/hoshizuna/
On the initial screen, there is a group of buttons at the top for selecting functions. When you access and load the page, the "Environment Edit" button is selected in red, and the Environment Edit screen is displayed in the upper right. The gray background area is where the 3D model is displayed. Normally, from this state, you will either add a model or edit the environment.
The initial screen immediately after accessing the application.
The initially displayed screen is called the "Control Screen." Pressing the "Dome View" button displays the "Dome View Screen" as a child window. The layout consists of these two screens. The Dome View is in dome master format (equidistant projection), so it can be projected directly into a planetarium or dome theater.
The left (1) is the initial Control Screen; the right (2) is the Dome View Screen.
Operating Environment
It works on WebGL-compatible web browsers, including smartphones and tablets that support DisplayPort Alt Mode . It supports control via keyboard, mouse, gamepad, and touch monitor.
When increasing the resolution or handling complex models, the computational load becomes high, so it is recommended to enable the browser's GPU acceleration. Basic operation has been confirmed on the following browsers: Chrome, Edge, Firefox, and Opera for Windows; Chrome for Chromebook; and Safari for Mac.
Compatible with various web browsers.
To project into a planetarium or dome theater, the dome's video system must be capable of external video input. Connect via an HDMI cable or similar, and input the Dome View screen in dome master format.
You can add models from the blue button at the top of the screen. Supported file formats include STL, a general-purpose 3D file format, and some OBJ files. You can also load SVG as 2D objects.
Button group at the top of the screen. Models can be loaded from the blue button on the left.
When loading, choose whether to use the original file's coordinates or place the center of gravity at the origin. Since the model's coordinates can be moved later, if you just want to load it, selecting "Move to origin and add" is fine. When loading multiple files whose positional relationships are predetermined, select "Add according to coordinates.”
Origin selection when loading a file.
If displayed locations overlap, a smaller model may be hidden behind a larger model. You can check hidden models by editing the coordinates, size, and transparency from the Model Edit screen described later.
Multiple models can be placed in the space.