Heightened Reality Games
Scenario 1: Your friend may be an evil impersonator.
Player inputs description earlier in the story.
This is broken down to keywords by our LLM for later use.
Later, the story asks player to confirm the friend via image.
Player snaps a photo or uploads image.
Onboard Machine Vision analyzes descriptive keywords and classifies objects based on their presence.
If a specific threshold of keywords match, this person is not an impersonator.