arXiv:2606. 29020v1 Announce Type: cross Abstract: Weather synthesis aims to add weather effects to input videos while preserving scene identity, structure, and motion.
Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv AI.
arXiv:2609.36810v1 Announce Type: new Abstract: Video world models aim to predict future content from an observed scene while following prescribed camera motion. Real-world scene evolution is determi...
arXiv:2606.20083v4 Announce Type: replace Abstract: Video world models are moving toward preserving an observed world under controllable camera and object motion while allowing its environmental stat...
The paper introduces Physically Plausible Video Generation (PPVG), a method that generates videos consistent with physical laws by treating physical evolution as a chain of causally connected events. It employs three modules: Physics-driven Event Chain Reasoning to decompose phenomena into scene-graph events, Transition-aware Routed Keyframe Conditioning to guide keyframe synthesis for smooth transitions, and Physics-injected Contrastive Semantic Guidance to steer generation toward plausible dynamics. Experiments on multiple physics benchmarks show improved physical plausibility compared to prior approaches.
Reliable perception under diverse weather conditions remains a major challenge for autonomous driving systems. A common strategy to improve robustness is either to synthesize adverse weather conditions for training perception models or to apply weather-removal techniques to recover clean inputs.
arXiv:2607. 07601v1 Announce Type: cross Abstract: Safety evaluation for autonomous driving is dominated by rare, safety-critical interactions, motivating simulators that can deliberately synthesize corner cases with photorealistic observations.
Driving World Models (DWMs) have recently advanced rapidly with generative models, yet most existing methods mainly focus on conditional scene generation and lack explicit 3D scene understanding, lang...