Running into issues with head-mounted tracking mode? While the Orion beta represents a massive step forward, it’s still possible to encounter tracking and framerate problems. Here’s a quick guide to troubleshooting the most common causes of poor tracking – the HMD policy flag, bandwidth throttling, and calibration.
Check the known issues
Go to developer.leapmotion.com/get-started and check the known issues for Orion. Our team is probably crushing the problem as we speak.
HMD policy flag
If you notice poor tracking in your app or development environment (such as Unity), make sure that the
POLICY_OPTIMIZE_HMD flag is activated – as this triggers the head-mounted tracking mode. In Unity, this can be achieved by checking the “Is Head Mounted” box. Check out our documentation for more details. It’s also important to ensure that the HandController object is appropriately oriented, so that it follows the OVRCamera within the scene.
Diagnosing framerate and processing time
Using the Diagnostic Visualizer, you can quickly assess your Leap Motion Controller’s performance. To access it, right-click on the Leap Motion taskbar icon and launch “Visualizer.” Once the program launches, hit H and L to show additional diagnostic data, and hit V to change to a top-down perspective.
A healthy setup will display device and data framerates running ~115 FPS and processing time ranging between 3-8 milliseconds.* If you see lower values, follow these steps:
- Make sure that your system fulfills our minimum system requirements for intensive VR apps.
- Check that you’re running the latest version of the Leap Motion software.
- Ensure that the Leap Motion Controller is not plugged into the DK2 USB port or an external hub.
- Sometimes too many high-bandwidth devices on a single host adaptor can cause throttling. Try moving your controller cord to a different USB port.
*These numbers reflect the current state of the software, as VR tracking has been evolving too quickly to be fully optimized. We expect to bring the latency down in the coming months.
If these steps don’t resolve your low framerate or high processing time, email us at email@example.com or post on the community forums and we’ll help troubleshoot. Post your system specs along with the diagnostic data.
Sometimes the Leap Motion sensors can be knocked out of their initial alignment, either due to shipping, from a fall, or other disturbance. This can cause:
- Persistent jumpiness
- Frequent discontinuities in the tracking data
- Aberrations in tracking data that occur only in certain areas of the field of view
- Poor tracking range
When this happens, a quick recalibration is in order! For VR tracking, we recommend calibration of 90% or higher. You can recalibrate your device from the Leap Motion Control Panel – check out our documentation for more details.
Note: While the calibration tool will indicate success at 80%, be sure to continue calibrating until you hit 90 or more. If you can’t achieve a high enough score with standard calibration, try the following:
- make sure that the front face of the Leap Motion Controller is perfectly clean
- use a very clean mirror or piece of glass
- make sure that you’re holding the Leap Motion Controller above the reflective surface before calibration starts (instead of sticking it into the FOV after it’s begun)
- start in the center and spiral outward to the edge, then back inward to ensure uniform coverage all at once (if you focus too much in one particular corner the score may go up despite making the calibration worse)
Orion features radical improvements in how we handle cluttered backgrounds and ambient light – so you can bring your hand close to, and even in contact with, other surfaces. The expanded range also means that you should be able to reach to your full arm extension without losing tracking.
Brightly lit or pure darkness – it usually doesn’t matter. The Leap Motion Controller works by emitting infrared light with cameras designed to see on that spectrum, then applying algorithms to the raw sensor data. Since the device’s cameras sync with its infrared LEDs, it can easily filter out virtually all environmental light, including fluorescent lighting and the light from your computer screen.
If it looks dark in infrared, it can’t interfere with tracking. Because the infrared data appears in greyscale, our tracking algorithms rely in part on the contrast between background objects (your desk, walls, etc.) and foreground objects (your hands).
If you’re still experiencing tracking issue, please email us at firstname.lastname@example.org or post on the community forums with your system specs, Diagnostic Visualizer data, and (if possible) a video from the VR Visualizer (accessible from the tray icon). Your feedback goes directly towards improving the platform, so we’d love to hear from you.