[SLAM] ***AR Tracking based on which tools?

SLAM虽然小复杂，但对于开发者而言，ar sdk通常会解决这个问题。

所以相对于识别，跟踪是个看上去高大上但实则不需要关注细节的部分。

识别，要技术深耕；跟踪，需行业深耕。

在此了解下常见的ar sdk with v-slam：

• 8 Best augmented reality SDK for AR development for iOS and Android in 2017
• TOP 10 AR SDK for iOS and Android Development

阅读笔记

---

• SDK Feature Comparison

	Vuforia	EasyAR	Wikitude	ARToolKit	Kudan	MaxST	Xzimg	NyARToolKit
Maximum distance capturing / holding marker (m)	1.2 / 3.7	0.9 / 2.7	0.8 / 3	3 / 3	0.8 / 3	0.5 / 0.9	0.7 /	0.7 / 1
Recognition stability of immovable marker	10	7	6	8	10	7	8	5
Recognition stability of movable marker	6	3	4	6	6	2	7	3
Minimum angle recognition	30	35	40	10	30	50	35	45
Minimum visibility for recognition overlapped marker	20%	10%	30%	100%	25%	50%	10%	75%
2D Recognition	✓	✓	✓	✓  (bordered)	✓	✓	✓	✓
3D Recognition	✓	–	✓ (beta)	–	✓	✓	–	–
Geo-Location	–	–	✓	–	–	–	–	–
Cloud Recognition	✓	–	✓	–	–	–	–	–
SLAM	–	–	✓	–	✓	✓	–	–
Total (rating)	7.1	4.4	7.5	2.8	6.9	5.2	4.7	3.1

EasyAR2.0有SLAM

ARToolKit 6 new features

Planar Tracker: Using the same algorithms in our image tracker, we can also track unprepared planes. Just similar to SLAM technology used by Apple ARKit.

【需测试】