English(EN)WorldSculpt: Generating Compositional Worlds from Grounded Videos
新框架通过3D映射和合成数据增强长时视频生成
作者PulseAugur 编辑部·[10 个来源]·
研究人员开发了几个新的框架和方法来生成长时、世界一致的视频。SolarWM为交互式视频世界模型提供了一个开放的基础，统一了高达33B参数模型的各种数据源和训练配方。另一种方法使用Unreal Engine创建合成的、动作条件化的视频数据，生成数千小时的高分辨率素材。此外，Statebench和Stateagent旨在通过关注世界状态推理来改进视频续写，而OctWorld和Streaming4D则利用3D映射和增量重建来实现长距离、一致的视频生成，并提高效率。
AI
arXiv:2602.14941v2 Announce Type: replace-cross Abstract: Maintaining spatial world consistency over long horizons remains a central challenge for camera-controllable video generation. Existing memory-based approaches often condition generation on globally reconstructed 3D scenes…
Adapting a single-object 3D generative prior to multi-view observations enables scalable compositional mesh reconstruction of densely cluttered scenes with severe occlusion.
We present OctWorld, a video diffusion framework with persistent 3D memory for generating explorable, world-consistent, and high-fidelity visual scenes. Given a single image, OctWorld performs stable autoregressive world generation along user-specified camera trajectories. We foc…
SolarWM provides an open framework and unified training recipe for building interactive video world models across diverse data sources and generator backbones, enabling long-horizon real-time rollouts.
arXiv:2609.05416v1 Announce Type: new Abstract: We study the problem of generating a compositional 3D representation of a cluttered scene containing hundreds of objects. The goal is to represent the scene as a collection of individual object meshes placed in a shared world frame,…
arXiv:2609.03919v1 Announce Type: new Abstract: We present OctWorld, a video diffusion framework with persistent 3D memory for generating explorable, world-consistent, and high-fidelity visual scenes. Given a single image, OctWorld performs stable autoregressive world generation …
arXiv:2609.03557v1 Announce Type: new Abstract: Action-conditioned video models require large-scale visual data paired with control signals that are temporally aligned with the resulting scene transitions. Such supervision is difficult to obtain from ordinary real-world video bec…
arXiv:2609.03673v1 Announce Type: new Abstract: Video generators build long videos by composing shorter parts, either by generating segments one after another or by autoregressively extending chunks. Each new part usually depends on memories of historical observations, such as re…
arXiv:2609.02886v1 Announce Type: new Abstract: We introduce SolarWM, a fully open foundation for building interactive video world models from data preparation through long-horizon inference. Training across heterogeneous data sources and video backbones is challenging: datasets …
arXiv:2609.00610v1 Announce Type: new Abstract: Current 4D generation paradigms are often bottlenecked by a sequential decoupling design: video is generated first, followed by 3D reconstruction, leading to high interaction latency. This limits applications in interactive real-tim…
