Geometry as Address: Routing Attention to Visual Memory for Long-Horizon Camera-Controlled Video Generation
cs.CV
Submitted: 2026-09-28
Updated: 2026-09-28
Comments: Project Page: https://zju3dv.github.io/geometry-as-address/
Project page: https://zju3dv.github.io/geometry-as-address
License: http://creativecommons.org/licenses/by/4.0/
The gist: Long-horizon camera-controlled video generation requires recovering previously observed content from an ever-growing visual history.
Terminology
Abstract
Long-horizon camera-controlled video generation requires recovering previously observed content from an ever-growing visual history. Existing approaches either search historical context implicitly or reconstruct it into persistent 3D memory, facing inefficient memory access or accumulated geometric errors. Our key insight is that geometry need not explain the scene--it only needs to determine where visual memory should be read from, while attention decides what should be recovered. Based on this insight, we introduce GEAR, a Geometry-Enabled Attention Routing framework that uses geometry as an explicit token-level address for visual memory. Rather than fusing historical observations into a persistent global 3D representation, GEAR retains them as frame latents and uses per-frame geometry only to establish token-level correspondences with target views, thereby avoiding persistent error accumulation from global fusion. Guided by these correspondences, Geometric Correspondence Attention (GCA) selectively injects geometrically matched historical features into noisy target patches during denoising. We further introduce an Invisible Octree to accumulate visibility evidence and reject geometrically plausible but occluded correspondences. Extensive experiments demonstrate that GEAR achieves state-of-the-art visual quality, precise camera control, and revisit consistency, enabling minute-long video generation along challenging trajectories.
Sources
- AlayaWorld: Interactive Long-Horizon World Modeling - Full Technical Report (v1.1)
- Qwen3-VL Technical Report
- Positional Encoding Field
- Matrix-game 2.0: An open-source, real-time, and streaming interactive world model
- RELIC: Interactive Video World Model with Long-Horizon Memory
- ViPE: Video Pose Engine for 3D Geometric Perception
- Advancing Open-source World Models
- Lyra 2.0: Explorable Generative 3D Worlds
- HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds
- Seedance 2.0: Advancing Video Generation for World Complexity
- Wan: Open and Advanced Large-Scale Video Generative Models
- Towards Accurate Generative Models of Video: A New Metric & Challenges
Related papers
- Loss Knows Best: Detecting Annotation Errors in Videos via Loss Trajectories
- AnchorWeave: World-Consistent Video Generation with Retrieved Local Spatial Memories
- Benchmarking the Robustness of Foundation Models for Mammography under Domain Shift
- MambaX-Net: Dual-Input Mamba-Enhanced Cross-Attention Network for Longitudinal MRI Segmentation
- TeleOCR: Navigating Document Parsing Across Digital and Camera-Captured Documents
- A Survey on Efficient Vision-Language-Action Models