The idea
Describe a camera move the morning after a party. AfterCam reconstructs a virtual path through the time and views the group recorded.
How it works
- Friends record a birthday, wedding, or event from different places around the room.
- After the event, describe or shape a path such as moving toward the cake and orbiting the candles.
- Reconstruct the requested move through the captured time, even if no phone filmed that exact path.
AI’s role: Translate a natural-language request into a camera path and time range; reconstruction remains constrained by captured angles and scene details.
First demonstration
Record one birthday moment with five phones. Request a path from behind the group toward the cake and around the subject, then compare the result with the actual coverage.
What to solve next
How should the interface show where a requested virtual move runs out of real captured information?