Google DeepMind has introduced a new AI-powered feature for Google Photos called Auto frame, which allows users to re-compose images after they have been taken. The system first estimates a 3D scene and camera perspective from the original photo. It then uses generative AI, specifically a latent diffusion model, to fill in any newly revealed areas and create an authentic new perspective of the scene. AI
Summary written by gemini-2.5-flash-lite from 1 source. How we write summaries →
IMPACT Enhances user control over photo composition post-capture, potentially reducing the need for manual editing or retakes.
RANK_REASON This is a feature update to an existing product, not a new model release or paradigm-shifting capability.
