TourPhysics: Bringing Physics to World Models for Exploration and Manipulation from a Single Image
Classification
View four quadrants- Major category
- Benchmarks & simulators
- Quadrant
- Not applicable
- Architecture
- Not applicable
- Prediction paradigm
- Not applicable
- Subcategories
- Neural world simulators
- Source review status
- Not assigned
Category review. A physics simulator executes user interventions and fixes trajectories before neural video synthesis, with persistent state and appearance memory; the work is an interactive neural/physics simulation framework, not an action policy. The cataloged contribution is a dataset/data-generation method or evaluation/simulation resource, not the architecture of an evaluated or external policy. Reading evidence
Contribution
TourPhysics renders persistent camera tours and object manipulations from a single image plus a declared physical scene. A simulator fixes each trajectory before diffusion generates its appearance; accepted windows publish physical endpoints and visual memory together. The strongest evidence concerns prescribed motion and modest improvements on held-out revisits, with limited physical-event support and no real-world validation.
Abstract
An abstract has not been added yet.
Affiliations
Fudan University; Institute of Artificial Intelligence (TeleAI), China Telecom; Nanjing University