Paper 2509.12201
OmniWorld: A Multi-Domain and Multi-Modal Dataset for 4D World Modeling
- Published
- Sep 2025
- Research lab
- Independent
- Citations
- 43
- GitHub
- 489 stars
01 In brief
Summary
OmniWorld is a large-scale, multi-domain, multi-modal dataset for 4D world modeling, introduced by Shanghai AI Lab and ZJU.
It comprises a self-collected OmniWorld-Game synthetic dataset (96K clips, 18.5M frames, 214+ hours) and curated public datasets from robot, human, and internet domains, totaling over 600K sequences and 300M frames.
OmniWorld provides annotations including depth maps, camera poses, text captions, optical flow, and foreground masks.
The OmniWorld-Game benchmark evaluates 3D geometric foundation models (e.g., DUSt3R, CUT3R, VGGT) and camera-controlled video generation models (e.g., AC3D, CamCtrl), revealing limitations of current SOTA methods in handling complex dynamics and long sequences.
Fine-tuning SOTA models on OmniWorld yields significant performance gains on public benchmarks (e.g., Sintel, KITTI, NYU-v2) and on OmniWorld-Game, validating its utility as a training resource.
The dataset aims to accelerate the development of general-purpose 4D world models.
02 From the paper
Abstract
The field of 4D world modeling - aiming to jointly capture spatial geometry and temporal dynamics - has witnessed remarkable progress in recent years, driven by advances in large-scale generative models and multimodal learning. However, the development of truly general 4D world models remains fundamentally constrained by the availability of high-quality data. Existing datasets and benchmarks often lack the dynamic complexity, multi-domain diversity, and spatial-temporal annotations required to support key tasks such as 4D geometric reconstruction, future prediction, and camera-control video generation. To address this gap, we introduce OmniWorld, a large-scale, multi-domain, multi-modal dataset specifically designed for 4D world modeling. OmniWorld consists of a newly collected OmniWorld-Game dataset and several curated public datasets spanning diverse domains. Compared with existing synthetic datasets, OmniWorld-Game provides richer modality coverage, larger scale, and more realistic dynamic interactions. Based on this dataset, we establish a challenging benchmark that exposes the limitations of current state-of-the-art (SOTA) approaches in modeling complex 4D environments. Moreover, fine-tuning existing SOTA methods on OmniWorld leads to significant performance gains across 4D reconstruction and video generation tasks, strongly validating OmniWorld as a powerful resource for training and evaluation. We envision OmniWorld as a catalyst for accelerating the development of general-purpose 4D world models, ultimately advancing machines' holistic understanding of the physical world.