Virtual Avatar Feature: System Requirements、Detailed Guide and Virtual Sound

In this Guide:

Computer Configuration Requirements
User Guide for the Virtual Avatar Feature
Virtual Sounds

Computer Configuration Requirements

Recommended computer configuration

OS	Windows 11 (or Windows 10)
System Version	64-bit version 1909 revision .1350 or later, or version 2004 and 20H2 revision .789 or later
CPU	Intel Core I7 12700K（or i5-13600）
Memory	32GB
GPU	Nvidia RTX 4070Super 12GB
Storage Space	Requires 20GB or more of available space
Remark	High-quality 60fps rendering frame rate

Mid-range computer configuration

OS	Windows 11 (or Windows 10)
System Version	64-bit version 1909 revision .1350 or later, or version 2004 and 20H2 revision .789 or later
CPU	Intel Core I5 12400
Memory	32GB
GPU	Nvidia RTX 3050 8GB
Storage Space	Requires 20GB or more of available space
Remark	Medium settings, 45fps rendering frame rate

Minimum computer configuration

OS	Windows 11 (or Windows 10)
System Version	64-bit version 1909 revision .1350 or later, or version 2004 and 20H2 revision .789 or later
CPU	Intel Core I5 12400
Memory	16GB
GPU	Nvidia GTX 1060 6GB
Storage Space	Requires 20GB or more of available space
Remark	Low quality 30fps rendering frame rate

Note:

Currently, this feature only supports NVIDIA platform graphics cards.
Please update your graphics card driver to the latest version before use.

NVIDIA GPU support list

Please use the following GPUs and ensure that the GPU driver version is the latest..

NVIDIA	RTX 5090
	RTX 5080
	RTX 5070
	RTX 5060
	RTX 5050
	RTX 4090
	RTX 4080
	RTX 4070
	RTX 4060
	RTX 4050
	RTX 3090
	RTX 3080
	RTX 3070
	RTX 3060
	RTX 3050
	RTX 2080
	RTX 2070
	RTX 2060
	RTX 2050
	GTX 1660
	GTX 1650
	GTX 1080
	GTX 1070
	GTX 1060

User Guide for the Virtual Avatar Feature

Highlight Features

★ Real-time, precise character motion capture paired with a virtual camera system that simulates realistic camera movements.

★ One-click switching between virtual avatars and scenes.

★ Intelligent AI tracking for rapid output of high-quality virtual content.

Setup and Usage Requirements

This section mainly includes Tiny 3 installation and venue attire.

Tiny 3 Placement Requirements

During use, please place the Tiny 3 on a horizontal table or tripod. Keep the Tiny 3 base as level as possible to ensure the virtual character remains in the correct position during AI tracking or gimbal rotation.
The Tiny 3 needs to be mounted at a suitable height to ensure that the subject's face and all parts of the body can be captured in the shot.
After placing the Tiny 3, measure the vertical distance between the Tiny 3 and the ground, accurate to the centimeter.

place the tiny 3

Note: Currently, this feature is only supported by OBSBOT Tiny 3 and Tiny 3 Lite products.

Clothing requirements

Avoid wearing masks or other items that obscure your face or any part of your body.
To avoid wearing overly loose clothing that could affect the accuracy of 3D motion capture.

clothing requirement

Note: Currently, this function only supports one person to perform 3D motion capture at the same time.

Function Introduction

Download resource package

1. Download the OBSBOT Center software and ensure that Tiny 3 or Tiny 3 Lite products are correctly connected.

download the obsbot center software

2. Click and open the virtual avatar function window on the main interface or video preview window of the OBSBOT Center software.

open the virtual avatar function window

3. Upon first use of this feature, clicking the "Start" button will automatically download the necessary resource package. Please wait patiently for the download to complete. To ensure a successful download, please run the OBSBOT Center software with administrator privileges.

first use of this feature

Create a character

After entering the [Virtual Avatar] function for the first time, please follow the prompts to complete the virtual character selection, virtual scene selection, and motion capture-related settings.

1. Virtual Character Selection: This version offers six different styles of virtual characters. Enter a custom character name and click "Next".

select the virtual character

2. Virtual Scene Selection: This version offers four virtual scenes. In addition, you can choose whether to display dynamic effects within the scene.

select the virtual scene

3. In the props tab, you can select a virtual character holding a prop.

select a virtual character holding a prop

Motion capture parameter settings

After selecting the virtual character and virtual scene, you will enter the motion capture settings interface.

enter the virtual setting interface

1. Motion capture camera selection: Choose the Tiny 3 or Tiny 3 Lite device for 3D motion capture.

2. Virtual timbre: Select the audio voice changer

Note: For first-time use of the [Virtual Voice] function, you need to download and install the third-party audio driver VB-CABLE, and add VB-CABLE to output the virtual voice in the third-party software.

3. Motion Capture Camera View: After opening, you can simultaneously monitor the virtual rendered screen and the original screen from the motion capture camera.

4. Tracking Parameter Settings: You can adjust the tracking sensitivity of the body skeleton (hands and legs) separately.

Lowering the tracking sensitivity will reduce the range and sensitivity of the virtual character's movements compared to a real person, but the overall movement will be smoother and more stable. Please customize the values according to the actual situation during testing.

5. Finger Tracking: You can adjust the tracking sensitivity parameters for the left and right hands separately.

6. Facial Tracking: You can turn virtual character facial tracking on or off as needed.

7. Calibration Parameter Settings: Includes real person height information and camera height parameters.

Height: To increase the accuracy of 3D motion capture, please fill in the height information of the real person being motion captured.
Camera Height Above Ground: Please enter the Tiny 3 camera height above ground measured in the steps above.

Note: After creating the virtual character, to ensure motion capture quality, please avoid moving the camera position during use.

avoid moving the camera position during use

Role Management

To manage one or more created roles, the management interface is generally divided into the following functional areas:

role management

1. Status Indicator Area: Displays the current motion capture camera occupancy status and 3D motion capture status in real time.

2. Quick Function Operations:

Motion Capture Camera View: When enabled, it allows simultaneous monitoring of the original camera view and the virtual rendered view.
Virtual Character Output: When enabled, allows direct monitoring, recording, and output control of the view in the Center preview window.

3.Character Management Area:

Supports custom modification of created virtual character names.
Allows one-click switching between the current virtual avatar and virtual scene content via the character management card.

Virtual screen preview monitoring

virtual screen preview monitoring

After enabling virtual character output, you can monitor the virtual avatar content in real-time through the video preview window of the Center software.
You can enable virtual camera output in the preview output window to capture the virtual image after 3D motion capture using third-party software.
You can also enable video recording in the preview output window to record and save the real-time 3D motion capture footage.
Clicking the "AI Settings" button will temporarily switch the video preview window to the original video feed from the current motion capture camera, allowing you to configure AI tracking, image settings, etc. However, the virtual camera output will still be the virtual image stream after 3D motion capture.

Frequently Asked Questions (FAQ)

Q1: Why is the virtual character moving erratically or not moving at all?

Please check if your Tiny 3 or Tiny 3 Lite device is properly connected and ensure the Center software can monitor the camera feed.
Please ensure that the camera can clearly capture the person's face and that there are no obvious obstacles between the person and the camera.
Please check if multiple people are in the frame simultaneously. Currently, this function only supports 3D motion capture for one person at a time.

Q2: The virtual character's motion capture effect is poor. How can I improve the 3D motion capture effect?

When starting and using this function for the first time, please ensure the entire character is in the frame to help the system complete the initialization, allowing for more accurate motion capture later.
Please ensure the camera gimbal base is level and parallel to the ground.
Please ensure the character's face is captured.
Avoid wearing masks or other items that obstruct the face and body.
Avoid wearing overly loose clothing, as this can affect the accuracy of 3D motion capture.
Ensure clean and bright lighting; do not use the device in dark or complex lighting environments.
Keep the background clean and tidy, avoiding images of people, posters, or other unwanted items.
Please try to lower the output resolution in the OBSBOT Center software for testing, for example, switch from the 60P mode to the 30P mode.

Q3: Why do virtual characters occasionally float in the air or sink into the ground?

Please note that when using the Avatar feature, keep a distance of more than 1 meter from the Tiny 3 camera.
Please check if the Tiny 3 tripod base is level with the ground during the installation process. Additionally, you can reset the Tiny 3 tripod to its initial state before enabling the function. Of course, after the function is enabled, you can freely rotate the tripod and activate features such as person tracking.

Q4: How to resolve issues such as software lag, frame drops, crashes, freezes, and unresponsiveness?

If the virtual avatar function experiences black screens, lag, frame drops, crashes, freezes, or unresponsiveness, please restart the virtual avatar module. If the problem persists, try closing other resource-intensive computer applications.
This version only supports NVIDIA platforms. Please update your graphics card driver to the latest version before use.

Q5: Why do I have two or more motion capture cameras with virtual avatars set up, but only one avatar can be displayed correctly at a time, while the others display the message "Virtual resources are being preempted"?

A virtual avatar can only process the effects of one camera and one character at a time.
When a virtual avatar motion capture camera is outputting or recording via virtual camera, processing of other virtual avatars is paused.
When a motion capture camera without a virtual avatar outputs and records virtual video, the camera video stream corresponding to the highlighted window occupies virtual resources, and processing of other virtual avatars is paused.