Microgestures expand the capabilities of hand tracking by recognizing low-calorie thumb tap and thumb swipe motions performed on the side of the index finger. These gestures trigger discrete D-pad-like directional commands. The hand pose and motion of the thumb is as follows: initially, the thumb must be raised above the index finger (not touching the index finger). The other fingers should be slightly curled as in the picture below for best performance: i.e. not too extended, nor completely curled into a fist. A tap is performed by touching the middle segment of the index finger with the thumb, and then lifting the thumb. The four directional thumb swipes performed on the surface of the index finger are:

Left swipe: a swipe towards the index fingertip on the right hand, and away from the index fingertip on the left hand. On the right hand for example, the motion is as follows: the thumb starts raised above the index finger, touches the middle segment of the index finger, slides towards the index fingertip, and lifts.

Right swipe: the same motion as the left swipe, but in the opposite direction. On the right hand for example, the thumb starts raised above the index finger, touches the middle segment of the index finger, slides away from the index fingertip, and lifts.

Forward swipe: the thumb starts raised above the index finger, touches the middle segment of the index finger, slides forward, and lifts.

Backward swipe: the thumb starts raised above the index finger, touches the middle segment of the index finger, slides backward/downward, and lifts. Note that the motions can be performed at moderate to quick speeds, and should be performed in one smooth motion. The detection of the gesture happens at the end of the motion.

Compatibility

Harware Compatibility

Quest 2, Quest Pro, Quest 3 Family of Devices

Software Requirements

Unreal Engine 5.5 or higher

Meta XR plugin V77 or higher

How do Microgestures work?

Microgestures in Unreal Engine are exposed via the Enhanced Input system, similar to how other system gestures - such as pinch - and controller inputs are exposed. To use microgestures in your project, you must set up Input Actions for each microgesture you want to support as well as an Input Mapping Context that maps those Input Actions to the microgesture input.

Creating an Input Action for a Microgesture

In the Content Drawer, select Add New > Input > Input Action to create the Input Action asset.

Add Input Action for left swipe on left hand

Name the new Input Action asset. In this example, it is named IA_Microgesture_L_Left_Swipe because this is the Input Action for the left swipe gesture on the left hand.

Name the Input Action for left swipe on left hand

Double-click the new Input Action asset in the Content Drawer to open it in the editor.

The microgesture inputs are either active or not, like a button press. Set the Action > Value Type to Digital (bool).

Creating the Input Mapping Context for Microgestures

In the Content Drawer, select Add New > Input > Input Mapping Context to create the Input Mapping Context asset.

Add Input Mapping Context for Microgestures

Name the new Input Mapping Context asset. In this example, it is named IMC_Microgestures because this is the Input Mapping Context for all microgestures.

Name Input Mapping Context for Microgestures

Double-click the new Input Mapping Context asset in the Content Drawer to open it in the editor.

Click the plus button next to Mappings to add a new mapping to the Input Mapping Context.

Select the Input Action you want to add a mapping for from the dropdown for the new mapping. In this example, it is IA_Microgesture_L_Left_Swipe.

Select Input Action for Left Swipt for left hand

Select the microgesture input you want to map the Input Action to from the dropdown for the new mapping. In this example, it is Oculus Hand (L) Microgesture - Swipe Left.

Select Input for Left Swipt for left hand

Repeat this process for the remaining microgestures you want to support in your project.

For more information on creating Input Actions and Input Mapping Context assets in Unreal Engine, see the Enhanced Input⁠ documentation.

Using the Input Mappings in your project

The Input Mapping Context asset must be added to the player through the Project Settings or via Blueprint so the Enhanced Input system knows to listen for those inputs. If you always want to use the Input Mapping Context in your project, you can add it to the Default Mapping Contexts in the Project Settings. If you only want to use the Input Mapping Context in certain circumstances, you can add it using Blueprint in response to some event. For example, it is common to add it to the Pawn when the Begin play event is fired so that specific Pawn type will always have the same input mappings. You could also use the Level Blueprint to add the Input Mapping Context to the player for a specific level that requires specialized inputs you want to have mapped to the Microgestures.

To add the Input Mapping Context through the Project Settings, open the Project Settings from the Edit menu and select Enhanced Input on the left. Click the plus button next to Default Mapping Contexts under the Enhanced Input section. Assign the Input Mapping Context you want to add to the Input Mapping Context property.

To add the Input Mapping Context through Blueprint, use the Add Mapping Context function. Connect a reference to the Enhanced Input Local Player Subsystem to the Target input pin and assign the Input Mapping Context you want to add to the Mapping Context input pin.

Reacting to the Microgesture Inputs

Each Input Action you created for a microgesture has a corresponding Input Action event that can be bound in Blueprint. This enables you to listen for each of the Microgestures and perform some action or series of actions in response. For example, you could bind the Thumb Tap input to open a menu, use the directional swipe inputs to navigate the menu, and use the Thumb Tap input again to select the menu item.

To bind the Input Action event in Blueprint:

Drag the Input Action asset from the Content Drawer into the Blueprint graph.

Connect nodes to the corresponding output pin, depending on when you want to listen for the Input Action event. For example, generally you will want to connect a node to the Started output pin to listen for when the Input Action is first detected. Similarly, if you want to repeatedly listen for the input while it is being detected, you can connect a node to the Triggered output pin.