WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory

Hugging Face Daily Papers Papers
video-generation world-model 3d-aware-memory camera-control implicit-memory academic-paper ai-research

Summary

WorldCrafter is a video world model that learns a camera-queryable implicit 3D-aware memory for consistent and camera-controllable streaming scene exploration from a single image or text prompt.

Video world models enable interactive exploration of dynamic environments, yet struggle to respect prior observations over long horizons and across viewpoints. We present WorldCrafter, a video world model that learns a camera-queryable implicit 3D-aware memory for this purpose. The key insight is to let the requested viewpoint shape how multi-view evidence is compressed into the video generator's limited token budget. Trained jointly with the video generator, a memory encoder and pose-conditioned readout module integrate historical observations into a fixed set of target view-specific tokens before denoising, without explicit depth-based correspondences. By combining this memory with recent temporal context and few-step distillation, WorldCrafter enables streaming scene exploration from a single input image or text prompt. Experiments across static and dynamic scenes show substantial gains in long-horizon consistency and camera-control accuracy while preserving visual quality during minute-scale exploration.
Original Article
View Cached Full Text

Cached at: 09/22/26, 03:24 AM

Paper page - WorldCrafter: Consistent Video World Model with Implicit 3D-aware Memory

Source: https://huggingface.co/papers/2609.24984 Authors:

,

,

,

,

,

,

,

,

,

Abstract

Videoworldmodelsenableinteractiveexplorationofdynamicenvironments,yetstruggletorespectpriorobservationsoverlonghorizonsandacrossviewpoints.WepresentWorldCrafter,avideoworldmodelthatlearnsacamera-queryableimplicit3D-awarememoryforthispurpose.Thekeyinsightistolettherequestedviewpointshapehowmulti-viewevidenceiscompressedintothevideogenerator’slimitedtokenbudget.Trainedjointlywiththevideogenerator,amemoryencoderandpose-conditionedreadoutmoduleintegratehistoricalobservationsintoafixedsetoftargetview-specifictokensbeforedenoising,withoutexplicitdepth-basedcorrespondences.Bycombiningthismemorywithrecenttemporalcontextandfew-stepdistillation,WorldCrafterenablesstreamingsceneexplorationfromasingleinputimageortextprompt.Experimentsacrossstaticanddynamicscenesshowsubstantialgainsinlong-horizonconsistencyandcamera-controlaccuracywhilepreservingvisualqualityduringminute-scaleexploration.

View arXiv pageView PDFProject pageGitHub2Add to collection

Get this paper in your agent:

hf papers read 2609\.24984

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper2

#### TencentARC/WorldCrafter-Fast Image-to-Video• Updated5 minutes ago • 3 #### TencentARC/WorldCrafter-Base Image-to-Video• 14B• Updated5 minutes ago • 2

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.24984 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.24984 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Wonder: Video World Model Done Better

Hugging Face Daily Papers

Wonder is a general-purpose video world model that enables real-time, camera-controllable world exploration from an image or conditional video. It introduces camera conditioning via dense coordinate fields, a sparse attention memory mechanism, and techniques to improve distillation, allowing minute-scale video generation at 16 FPS.

World in World: Explore the World with World Models

Hugging Face Daily Papers

The paper presents World in World, a training-free interface that enables flexible camera and time control in frozen autoregressive video world models by using correspondence-guided queries and evidence-wise attention guidance.

Latent Spatial Memory for Video World Models

Hugging Face Daily Papers

This paper introduces latent spatial memory for video world models, storing 3D scene information directly in diffusion latent space to avoid costly pixel-space reconstruction. The proposed Mirage framework achieves up to 10.57x faster generation and 55x memory reduction while achieving state-of-the-art performance on WorldScore and RealEstate10K.

ReWorld: An Interactive World Model with Long-Horizon Memory

Hugging Face Daily Papers

ReWorld presents an interactive world model that decouples short-horizon control from long-horizon memory using mixed attention windows and distribution-matching LoRA distillation, enabling real-time video generation with high action fidelity and recall.