Table of Contents

Mega mapping service

Spatial mapping is the core of Mega cloud services, serving as the central hub in the entire Mega workflow. It processes collected data from the physical world and outputs digital assets required for application development.

The following illustrates the position of spatial mapping in the overall Mega workflow:

flowchart LR  

A[Data collection] --> B>Build Mega Block map] --> C[Mega localization]

Feature introduction

EasyAR Mega adopts a fully cloud-based automated processing pipeline. Developers do not need to run any complex mapping algorithms locally. Simply upload compliant video data, and the cloud cluster will automatically complete the following tasks:

  • Feature extraction: Identify key visual features in the environment (e.g., corners, textures).

  • Spatial reconstruction: Calculate the 3D spatial coordinates of feature points to generate spatial data.

  • Loop closure detection: Automatically identify loop closures in the path to eliminate accumulated errors.

  • Georeferencing: Align the local coordinate system to the global geographic coordinate system using GNSS data.

  • Multi-map fusion: Support merging multiple independent collection blocks into a complete map (for large-scale scenes).

Block mapping process

Organize mapping tasks by creating and managing mapping projects. After successful mapping, a Block is generated.

flowchart LR  

A[Mapping project] --> B[Mapping task] --> C[Mapping result Block]
  • Mapping project: Used to organize and manage related mapping tasks
  • Mapping task: Performs specific mapping operations
  • Block: The final product of successful mapping

Different types of mapping tasks

Depending on your scene size and collection strategy, Mega provides four types of mapping tasks. Please create the corresponding type of mapping task according to the actual situation:

Regular single-map task

  • Applicable scenarios: Small-to-medium scale scenes, single collection covering the entire scene
  • Typical cases: Single-floor shopping malls, restaurants, single-floor offices, exhibition halls, parks, etc.
  • Collection device: GoPro Max 1st-gen 360-degree camera
  • Steps: Create a regular single-map task

Large-scale multi-image fusion task

  • Applicable scenarios: Ultra-large-scale scenes, multi-floor, multi-region connection scenarios
  • Typical cases: Large shopping malls (connecting floors), commercial districts (connecting indoor and outdoor), university campuses (connecting multiple buildings), etc.
  • Capture device: GoPro Max 1st-gen 360-degree camera
  • Steps: Create a large-scale multi-image fusion task

Small-scale mobile collection task

Mobile object capture task

  • Applicable scenarios: Single object
  • Typical cases: Exhibits, equipment, products, etc.
  • Capture devices: iPhone / ARCore Android phone
  • Steps: Create a mobile object capture task

Status description of mapping

In the Block mapping interface of the development center, you can see the real-time status of mapping tasks. Understanding these statuses helps you determine task progress:

Mapping Status Description Recommended Action
Under review Manual mapping video review in progress Wait for the review to complete
In queue Waiting for idle server resources Wait for server allocation to start processing
Generating Server processing in progress Wait for mapping to complete
Task completed Mapping completed View mapping results
Generation failed Mapping failed Troubleshoot

View mapping results

After the mapping is completed, you can further check the mapping results to confirm whether they meet the expected requirements.

Troubleshooting mapping failures

When a mapping task fails to generate, it is recommended to troubleshoot in the following ways:

  • Check the mapping report: View the mapping report and find solutions based on the report information.
  • Refer to troubleshooting guide: Consult the mapping failure troubleshooting guide for solutions to common issues.
  • Contact technical support: If the above methods do not resolve the problem, please contact EasyAR staff and provide the JobID from the mapping report for quick issue identification.

Next steps