Persistent Planar Memory for Video World Models

Hugging Face Daily Papers Papers
video-world-models 3d-memory video-diffusion planar-geometry depth-estimation autoregressive-generation

Summary

PPMem is a video world model that uses compact plane primitives as persistent 3D memory, jointly generating long coherent videos with explicit metric depth, normals, and exportable surfaces while outperforming prior 3D-grounded world models on DL3DV and RealEstate10K.

Video world models that generate and explore environments at scale need a persistent 3D memory of what they have produced. Plane primitives offer a natural candidate: a compact set of oriented planes captures the dominant structural elements (e.g., walls, floors, facades) of indoor and outdoor scenes, while requiring only 7 parameters per primitive. We present PPMem, a video world model that adopts plane primitives as its persistent 3D memory. PPMem augments a video diffusion transformer with a planar prediction head and a Depth Token Fusion (DTF) module: each autoregressive chunk jointly produces RGB video and planar geometry, which merges into a persistent plane memory that conditions subsequent chunks via rendered depth. The resulting memory is three orders of magnitude more compact than per-pixel point clouds (~100K primitives, 5MB for a full scene), geometrically explicit (metric depth, normals, exportable surfaces), cross-view consistent (jointly predicted by the shared backbone), and adaptive (refined by the merging procedure as generation proceeds). On DL3DV and RealEstate10K, PPMem outperforms prior 3D-grounded world models in video quality while jointly producing metric-scale depth, surface normals, and exportable planar geometry, sustaining coherence across hundreds of frames under large viewpoint changes. Project page: https://hyzcluster.github.io/ppmem.
Original Article
View Cached Full Text

Cached at: 10/19/26, 03:27 AM

Paper page - Persistent Planar Memory for Video World Models

Source: https://huggingface.co/papers/2610.23795

Abstract

PPMem is a video world model that uses compact plane primitives as persistent 3D memory to generate long coherent videos with explicit metric geometry.

Video world models that generate and explore environments at scale need a persistent 3D memory of what they have produced.Plane primitivesoffer a natural candidate: a compact set of oriented planes captures the dominant structural elements (e.g., walls, floors, facades) of indoor and outdoor scenes, while requiring only 7 parameters per primitive. We present PPMem, a video world model that adoptsplane primitivesas its persistent 3D memory. PPMem augments avideo diffusion transformerwith aplanar prediction headand aDepth Token Fusion(DTF) module: eachautoregressive chunkjointly produces RGB video and planar geometry, which merges into apersistent plane memorythat conditions subsequent chunks viarendered depth. The resulting memory is three orders of magnitude more compact than per-pixelpoint clouds(~100K primitives, 5MB for a full scene), geometrically explicit (metric depth, normals, exportable surfaces),cross-view consistent(jointly predicted by the shared backbone), and adaptive (refined by the merging procedure as generation proceeds). On DL3DV and RealEstate10K, PPMem outperforms prior 3D-grounded world models in video quality while jointly producing metric-scale depth,surface normals, and exportable planar geometry, sustaining coherence across hundreds of frames under large viewpoint changes. Project page: https://hyzcluster.github.io/ppmem.

View arXiv pageView PDFProject pageGitHub1Add to collection

Get this paper in your agent:

hf papers read 2610\.23795

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

Models citing this paper1

#### Robbyant-Research/PPMem Image-to-Video• 6B• Updatedabout 1 hour ago

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2610.23795 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2610.23795 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

Addressable Memory for Video World Models

Hugging Face Daily Papers

This paper introduces WorldTrace, a training-free memory framework for long-horizon video world models that keeps compressed cache addressable, plus LoopBench, a benchmark for episodic recall after long detours. It improves temporal consistency by +15.5% and episodic recall by +19.5% on LoopBench.

Honeycomb: Constant-Size Scene Memory Representation for Video World Models

Hugging Face Daily Papers

The paper introduces Honeycomb, a video world model that uses HexMemory, a low-rank fixed-size scene memory represented as six spatial and spatiotemporal planes, enabling constant storage during long-horizon video generation with strong generation quality and revisit consistency on WorldScore and RealEstate10K benchmarks.

Memory in Video World Models (6 minute read)

TLDR AI

A best-paper research from NVIDIA and collaborators introduces WorldTrace, a training-free framework that keeps compressed memory addressable in autoregressive video world models by assigning fixed slot-rank positions, enabling coherent long rollouts and long-range recall beyond the training horizon.

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.