Building Rich UI for MR in MRTK3
summary
MRTK3 replaces prefab hell with a hybrid Canvas that combines Unity’s RectTransform layout (resizable grids, anchors, 24/32 mm design units) with full 3D volumetric buttons, rounded-corner clipping and pixel‑perfect collider clipping; one button prefab now covers every variant (one prefab to rule them all). Visual feedback is authorable without code via a StateVisualizer low-code system, and a proxy interactor maps gamepad/mouse events into XRI so desktop, WebGL and HoloLens input all work; heavier 3D visuals cost some rendering performance.
AI-written summary · about
chapters (25)
- 0:00Intro
- 1:18New MR Design Language
- 9:33New UI
- 10:08Topic Sentence
- 10:30MRTK2 UI
- 11:14Pain Points
- 11:51Prefab Hell
- 12:39Unity Canvas
- 13:51Dynamic Layouts
- 14:30Hybrid Canvas
- 15:49Unitless Design Units
- 17:08Flex Ratios
- 18:17Reflow
- 18:37Grid Layout
- 19:18Clipping Scrolling
- 21:33State Visualizer
- 22:02Prefab Inheritance
- 23:34Effects
- 24:41Cross Platform Support
- 25:42How it Works
- 28:36Layout Groups
- 31:05Canvas Root Node
- 32:29Scaling
- 37:01Recap
- 37:46Questions
related talks
10mReality Proxy: Fluid Interactions with Real-World Objects in MR via Abstract Representations
17mContext-Aware Online Adaptation of Mixed Reality Interfaces
09mAdvanced Augmented Reality interactions
36mMixed Reality - 3 Axes of Improvement
50mAdvanced Mixed Reality Development and Best Practices
46mHoloLens 301: Advanced Mixed Reality Development and Best Practices
35mReactive Reality
44mShifting gears: How to win with AR