Google AI Research is developing a system to re-compose existing photos by altering camera angles. The model synthesizes missing perspectives to shift the viewer's vantage point without losing image fidelity. This incremental update to generative vision tools allows for precise framing adjustments. Practitioners can now automate composition fixes that previously required manual editing or reshooting.