Table of Contents

Mega tracking anomaly troubleshooting guide

Mega employs cloud-device fusion technology, combining cloud-based localization with device SLAM to maintain continuous tracking, ensuring AR content precisely anchors to the real world. Ideal tracking should exhibit no drift, low jitter, and high smoothness. This document focuses on tracking anomalies during Mega localization, helping you understand expectations, classify common errors, recognize device impacts, and determine when to contact EasyAR technical support.

Expected effect

Tracking anomalies typically stem from mapping data quality, environmental factors, or device compatibility. You can use MegaToolbox for troubleshooting. If issues involve runtime application effects, you may also directly test with your application.

With high-quality Mega Blocks and favorable environments, tracking should meet these standards:

  • Stable anchoring: After successful localization, virtual content remains fixed in world coordinates without noticeable drift.
  • Robust recovery: Automatically recovers tracking after brief occlusion, rapid rotation, or lighting changes.
  • Smooth performance: Content renders normally without significant jitter.
  • Environmental limitations: Tracking degrades near plain walls, empty floors, glass, or highly reflective surfaces.

If your experience aligns with the above, tracking is essentially normal.

Normal phenomena: Minor jitter or temporary loss is acceptable, especially during creation phase validation. Prolonged stability is considered normal.

Self-diagnosis using Mega Toolbox

Tracking anomalies usually occur after successful localization. If anomalies appear in your developed application, first check localization status in Mega Toolbox (refer to Mega Toolbox operation manual - On-site localization testing & localization issue feedback data recording tool). If results differ, verify your implementation.

Operation steps

  1. Install Mega Toolbox on your test device.
  2. Open Toolbox, navigate to On-site localization testing & localization issue feedback data recording, log in with your credentials.
  3. Select SLAM method and localization library.
  4. Bring device to exact problem location, observe Toolbox display.

For localization library configuration and on-site testing procedures, refer to Previewing actual runtime effects.

Interpretation based on Toolbox display

Mega Toolbox serves as a primary diagnostic tool. This standalone app helps determine whether issues originate from Mega or your application.

  • Localization anomaly

    When Toolbox displays statuses other than Found or NotFound, consult the Localization failure handling guide.

  • Successful localization

    Toolbox displays Found status with properly rendered content.

    Normal occurrences include:

    • Content aligns with scene, showing slow movement over time
    • Virtual content snaps back when exceeding certain deviation range

    Anomalous occurrences include:

    • Persistent unidirectional drifting
    • Cyclic drifting and snapping back
    • Progressive scaling (enlarging/shrinking)
    • Content jumping between multiple locations

Error classification and troubleshooting

Using Toolbox test results, match anomalies to corresponding recommendations:

Anomaly classification Possible causes Recommended actions
Persistent NotTracking status Featureless environment, poor lighting, or unsupported device Differentiate environment vs device factors by testing alternative areas or devices
Unidirectional drifting Device limitations causing SLAM drift Try higher-performance devices
Cyclic drift-correction Device limitations causing SLAM drift Test with superior devices
Progressive scaling Device limitations causing SLAM drift Switch to better-performing devices
Position jumping Similar scene areas causing localization interference Initiate localization from distinctive areas or use Bluetooth assistance

Further testing

For deeper investigation, conduct these additional tests. Developers should reference Mega supported devices and platform applications.

Testing with higher-end devices

Tracking performance varies across devices due to camera quality, sensor precision, and processing power. During creation validation, test across multiple device models to ensure compatibility. If tracking works normally on ARCore/ARKit devices, the issue is device-specific. For lower-performance devices, experiment with different SLAM methods.

Selecting alternative SLAM methods

Lower-end devices often suffer from poor SLAM precision due to sensor limitations and computing constraints. In Toolbox's On-site localization testing, test different SLAM options: FiveDof, ThreeDof, or ZeroDof to evaluate varying tracking effects.

DoF selection

Scenarios requiring feedback

While most tracking anomalies stem from device limitations, report these conditions to EasyAR support as they may involve Block algorithms or service issues:

  • Persistent frequent drift across multiple devices and locations
  • Toolbox fails to localize or shows abnormal localization
  • Significant performance variation under different weather/lighting conditions unresolved by supplemental updates

If issues persist, provide problem description, device information, screen recording, EIF file, and Mega service details to EasyAR. Before reporting, confirm the EIF reproduces the issue in Mega Studio.

This guide enables rapid tracking anomaly troubleshooting during creation, enhancing overall Mega Block quality.