Mira-Scene: Pixel-Aligned Layouts for Generative 3D Scene

Hugging Face Daily Papers Papers
3d-scene layout-generation canonical-coordinate-map diffusion-transformer pixel-aligned compositional-reconstruction generative-model

Summary

Mira-Scene introduces a compositional 3D scene reconstruction framework using pixel-aligned canonical coordinate maps for accurate object layouts, achieving significant improvements in layout accuracy over existing methods.

Single-image 3D object generation can now produce high-fidelity assets, yet accurately placing them into a coherent scene layout remains an open challenge. A central difficulty lies in how object layout is represented. Holistic methods absorb placement into a scene-level generation process, sacrificing object-level detail. Compositional methods preserve object fidelity by decoupling geometry from layout, but typically parameterize layout as sparse, unbounded pose variables that are difficult to learn and generalize poorly under scarce scene-level supervision.We present Mira-Scene, a compositional 3D scene reconstruction framework that replaces sparse pose regression with dense, bounded correspondence recovery. At its core is the Canonical Coordinate Map (CCM), a pixel-aligned field that maps each visible object pixel to a surface coordinate in the object's bounded canonical space. When paired with a scene-space Point Cloud Map (PCM) from monocular geometry estimation, CCM induces dense canonical-to-scene correspondences from which object transformations are recovered through robust geometric alignment. Because CCM operates in bounded canonical space, it provides a stable prediction target that can be trained from scalable object-level 3D data without requiring scene-level layout annotations. Mira-Scene further introduces a multimodal diffusion transformer that jointly generates object geometry and CCMs, using modality-specific expert streams with shared attention and positional encoding to promote geometry-layout consistency. Experiments on indoor, outdoor, synthetic, and in-the-wild scenes show that Mira-Scene substantially outperforms strong baselines in layout accuracy, achieving relative gains of 39.8% in 3D-IoU and 16.5% in 2D-IoU over SAM3D, using limited open-source training data.
Original Article
View Cached Full Text

Cached at: 09/22/26, 07:25 AM

Paper page - Mira-Scene: Pixel-Aligned Layouts for Generative 3D Scene

Source: https://huggingface.co/papers/2609.23796

Abstract

Single-image3Dobjectgenerationcannowproducehigh-fidelityassets,yetaccuratelyplacingthemintoacoherentscenelayoutremainsanopenchallenge.Acentraldifficultyliesinhowobjectlayoutisrepresented.Holisticmethodsabsorbplacementintoascene-levelgenerationprocess,sacrificingobject-leveldetail.Compositionalmethodspreserveobjectfidelitybydecouplinggeometryfromlayout,buttypicallyparameterizelayoutassparse,unboundedposevariablesthataredifficulttolearnandgeneralizepoorlyunderscarcescene-levelsupervision.WepresentMira-Scene,acompositional3Dscenereconstructionframeworkthatreplacessparseposeregressionwithdense,boundedcorrespondencerecovery.AtitscoreistheCanonicalCoordinateMap(CCM),apixel-alignedfieldthatmapseachvisibleobjectpixeltoasurfacecoordinateintheobject’sboundedcanonicalspace.Whenpairedwithascene-spacePointCloudMap(PCM)frommonoculargeometryestimation,CCMinducesdensecanonical-to-scenecorrespondencesfromwhichobjecttransformationsarerecoveredthroughrobustgeometricalignment.BecauseCCMoperatesinboundedcanonicalspace,itprovidesastablepredictiontargetthatcanbetrainedfromscalableobject-level3Ddatawithoutrequiringscene-levellayoutannotations.Mira-ScenefurtherintroducesamultimodaldiffusiontransformerthatjointlygeneratesobjectgeometryandCCMs,usingmodality-specificexpertstreamswithsharedattentionandpositionalencodingtopromotegeometry-layoutconsistency.Experimentsonindoor,outdoor,synthetic,andin-the-wildscenesshowthatMira-Scenesubstantiallyoutperformsstrongbaselinesinlayoutaccuracy,achievingrelativegainsof39.8%in3D-IoUand16.5%in2D-IoUoverSAM3D,usinglimitedopen-sourcetrainingdata.

View arXiv pageView PDFProject pageGitHub8Add to collection

Get this paper in your agent:

hf papers read 2609\.23796

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

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2609.23796 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2609.23796 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2609.23796 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

Pixal3D: Pixel-Aligned 3D Generation from Images

Hugging Face Daily Papers

Pixal3D introduces a pixel-aligned 3D generation approach that improves fidelity by establishing direct pixel-to-3D correspondences through back-projection conditioning, addressing issues in canonical space generation.

PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space

Hugging Face Daily Papers

PixWorld presents a unified pixel-space diffusion approach for 3D scene reconstruction and generation, overcoming limitations of latent-space methods by using direct image-level supervision and geometry-aware feature alignment. The method outperforms prior generation methods and matches state-of-the-art reconstruction methods.