Adobe’s Wonder turns images into a camera-controllable video world model
adobe · hf · 2026-07-29
Adobe Research and Johns Hopkins present Wonder, a general-purpose video world model for real-time, camera-controllable exploration. Starting from an image or a conditional video, it builds a playable world where users can move the camera, revisit explored regions, and discover unseen areas over long horizons.
Key ingredients include:
- Dense coordinate-field camera conditioning to turn camera motion into visual evidence.
- Sparse-attention memory for fast retrieval over long generation contexts.
- Distillation fixes that help the student model obey control signals while keeping diversity and long-term memory.
The system can synthesize minute-scale videos at 16 FPS and also supports video-conditioned re-shooting of existing dynamic scenes in real time.
More from Multimodal
- Nokia vs iPhone, car edition: AI image shows Tesla Roadster hovering over a Porsche — vaibhavbetter · 2026-10-16
- Midjourney multi-sref weight mixing recipe for cinematic light & motion portraits — michaelrabone · 2026-10-16
- MiniCPM-V 4.7 35B-A3B hits 17+ tokens/s on iPad with ~3GB memory — techNmak · 2026-10-16
- turbopixel: layer-and-mask generative compositing runs a 9B model on 4GB GPU — 3unjee · 2026-10-16
- Manga Screen: open-source ComfyUI nodes turn illustrations into printed manga panels — KnowledgeBig4364 · 2026-10-16
- AI re-cut of Game of Thrones' Battle of the Bastards stars an orange cat — AxonkaiLab · 2026-10-16