Augmented Reality Basics
summary
Augmented reality depends on computer vision not hardware hype and uses three concrete anchoring methods: image targets, face meshes, and anchors from feature points. Image targets match a stored image and fail if occluded; face meshes place virtual geometry on detected facial landmarks; feature‑point anchors build a sparse point cloud, run localization and plane estimation, create a signature cluster, and can be uploaded to the cloud for relocalization. Limitations include meter‑scale drift, occlusion sensitivity, and most phone AR feeling gimmicky versus costly headsets.
AI-written summary · about
chapters (9)
- 0:00Intro
- 0:34Extended Reality (XR)
- 4:32Image Targets
- 5:08Face Mesh
- 5:40Anchors From Feature Points
- 6:42Look for feature points
- 7:18Localization and Mapping
- 7:55Estimate Planes
- 8:33Ready to Place Anchor
related talks
01h 16mAugmented Reality For The Masses
44mDiscovering Augmented Reality with ARCore
18m4 Keys to Augmented Reality’s Future
29mExtend your reality with AR
42mHow to Augment Your Reality
37mExploring Augmented Reality with ARCore
40mDiscovering Augmented Reality with ARCore
56mAugmented Reality - The State of Play