PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space
Summary
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.
View Cached Full Text
Cached at: 07/07/26, 06:42 AM
Paper page - PixWorld: Unifying 3D Scene Generation and Reconstruction in Pixel Space
Source: https://huggingface.co/papers/2607.05373
Abstract
PixWorld presents a unified pixel-space diffusion approach for 3D reconstruction and generation that overcomes limitations of latent-space methods through direct image-level supervision and geometry-aware feature alignment.
3D reconstructionand generation are commonly tackled by separate paradigms: pixel-based regression for reconstruction, and latent diffusion for generation. Recent works attempt to unify them inlatent space, but with notable drawbacks: the diffusion objective is defined on latent features rather than the underlying 3D representation, and both branches suffer from information loss introduced by latent encoding, while requiring a pretrainedVariational Autoencoder(VAE) orRepresentation Autoencoder(RAE). In this paper, we reformulate these two tasks under a unified pixel-space diffusion paradigm and introduce PixWorld, a single model that jointly addresses3D reconstructionand generation. By supervising diffusion directly onrendered images, PixWorld removes the above limitations and aligns optimization with3D scene fidelity. Beyond photometric and perceptual supervision that operates at the 2D image level and lacks 3D geometric awareness, we further introduce ageometry perception lossthat aligns rendered views with their ground truth in the geometry-aware feature space of a pretrained3D foundation model, providing 3D structural supervision. PixWorld consistently outperforms prior latent-space generation methods and matches state-of-the-art reconstruction methods, demonstrating the superiority of a unified pixel-space approach.
View arXiv pageView PDFProject pageAdd to collection
Get this paper in your agent:
hf papers read 2607\.05373
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/2607.05373 in a model README.md to link it from this page.
Datasets citing this paper0
No dataset linking this paper
Cite arxiv.org/abs/2607.05373 in a dataset README.md to link it from this page.
Spaces citing this paper0
No Space linking this paper
Cite arxiv.org/abs/2607.05373 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
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.
World Tracing: Generative Pixel-Aligned Geometry Beyond the Visible
World Tracing introduces a generative pixel-aligned geometry representation that predicts 3D points aligned with observed pixels while completing occluded surfaces. It uses a diffusion transformer trained with pixel-space flow matching, achieving strong performance on visible-surface reconstruction and complete geometry generation across object, scene, and dynamic benchmarks.
TencentARC/Pixal3D
Pixal3D is a high-fidelity single-image-to-3D model by TencentARC and Microsoft, which explicitly lifts pixel features into 3D via back-projection for near-reconstruction-level geometry and PBR textures. The model is accepted to SIGGRAPH 2026, with inference code and demo available.
GenRecon: Bridging Generative Priors for Multi-View 3D Scene Reconstruction
GenRecon introduces a method for 3D scene reconstruction that integrates generative 3D priors with multi-view image conditioning, achieving high-fidelity, editable mesh reconstructions of indoor environments and outperforming existing methods by 16%.
HY-World 2.0: A Multi-Modal World Model for Reconstructing, Generating, and Simulating 3D Worlds
HY-World 2.0 is a multi-modal world model framework that generates high-fidelity 3D Gaussian Splatting scenes from text, images, and videos through specialized modules for panorama generation, trajectory planning, and scene composition, achieving state-of-the-art performance among open-source approaches.