Paper page - AnyRecon: Arbitrary-View 3D Reconstruction with Video Diffusion Model

Source: https://huggingface.co/papers/2604.19747

Abstract

AnyRecon enables scalable 3D reconstruction from arbitrary sparse inputs using diffusion models with persistent scene memory and geometry-aware conditioning for improved geometric consistency.

Sparse-view 3D reconstructionis essential for modeling scenes from casual captures, but remain challenging for non-generative reconstruction. Existing diffusion-based approaches mitigates this issues by synthesizing novel views, but they often condition on only one or two capture frames, which restricts geometric consistency and limits scalability to large or diverse scenes. We propose AnyRecon, a scalable framework for reconstruction from arbitrary and unordered sparse inputs that preserves explicit geometric control while supporting flexible conditioning cardinality. To support long-range conditioning, our method constructs a persistentglobal scene memoryvia a prependedcapture view cache, and removestemporal compressionto maintain frame-level correspondence under large viewpoint changes. Beyond better generative model, we also find that the interplay between generation and reconstruction is crucial for large-scale 3D scenes. Thus, we introduce ageometry-aware conditioningstrategy that couples generation and reconstruction through an explicit3D geometric memoryand geometry-driven capture-view retrieval. To ensure efficiency, we combine 4-stepdiffusion distillationwithcontext-window sparse attentionto reduce quadratic complexity. Extensive experiments demonstrate robust and scalable reconstruction across irregular inputs, large viewpoint gaps, and long trajectories.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2604\.19747

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

Models citing this paper1

#### Yutian10/AnyRecon Updatedabout 3 hours ago

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2604.19747 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2604.19747 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.