Differences between EasyAR Dense Spatial Map and ARKit/ARCore

This chapter introduces the differences between EasyAR Dense Spatial Map and similar ARKit/ARCore features, such as generating dense meshes or depth maps, and compares the advantages and disadvantages of using meshes and depth maps for realistic augmented reality effects.

Meshes generated by EasyAR and ARKit

EasyAR Dense Spatial Map reconstructs environment meshes in real time for occlusion, collision, and similar effects. ARKit provides a similar scene reconstruction feature and can export environment meshes. In principle, ARKit uses an RGB camera and a depth LiDAR camera, so it only supports devices after the fourth-generation iPad Pro. EasyAR Dense Spatial Map uses only an RGB camera, so it supports a wider range of device models.

Implement realistic AR effects: use meshes or depth maps

Google ARCore does not support generating mesh information for the environment. Instead, it can implement occlusion by providing a depth map for each frame.

Apple ARKit can obtain the depth of the rear camera and scene through a mechanism similar to depth map (SceneDepth), enabling collision and occlusion effects.

Both depth maps and meshes can be used in augmented reality to implement realistic occlusion, collision, and similar features. Each has advantages and disadvantages. The comparison is as follows:

	Depth map	Mesh
Representation dimension	2.5D	3D
Update method	Real-time update per frame	Incremental or batch reconstruction
Dynamic object support	Good	Poor
Cross-view consistency	Poor	Good
Persistence capability	Not supported	Supported