ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: de Lutio, Riccardo, Fischer, Tobias, Chang, Yen-Yu, Zhang, Yuxuan, Wu, Jay Zhangjie, Ren, Xuanchi, Shen, Tianchang, Tothova, Katarina, Gojcic, Zan, Turki, Haithem
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866910193808310272
author de Lutio, Riccardo
Fischer, Tobias
Chang, Yen-Yu
Zhang, Yuxuan
Wu, Jay Zhangjie
Ren, Xuanchi
Shen, Tianchang
Tothova, Katarina
Gojcic, Zan
Turki, Haithem
author_facet de Lutio, Riccardo
Fischer, Tobias
Chang, Yen-Yu
Zhang, Yuxuan
Wu, Jay Zhangjie
Ren, Xuanchi
Shen, Tianchang
Tothova, Katarina
Gojcic, Zan
Turki, Haithem
contents Per-scene optimization methods such as 3D Gaussian Splatting provide state-of-the-art novel view synthesis quality but extrapolate poorly to under-observed areas. Methods that leverage generative priors to correct artifacts in these areas hold promise but currently suffer from two shortcomings. The first is scalability, as existing methods use image diffusion models or bidirectional video models that are limited in the number of views they can generate in a single pass (and thus require a costly iterative distillation process for consistency). The second is quality itself, as generators used in prior work tend to produce outputs that are inconsistent with existing scene content and fail entirely in completely unobserved regions. To solve these, we propose a two-stage pipeline that leverages two key insights. First, we train a powerful bidirectional generative model with a novel opacity mixing strategy that encourages consistency with existing observations while retaining the model's ability to extrapolate novel content in unseen areas. Second, we distill it into a causal auto-regressive model that generates hundreds of frames in a single pass. This model can directly produce novel views or serve as pseudo-supervision to improve the underlying 3D representation in a simple and highly efficient manner. We evaluate our method extensively and demonstrate that it can generate plausible reconstructions in scenarios where existing approaches fail completely. When measured on commonly benchmarked datasets, we outperform all existing baselines by a wide margin, exceeding prior state-of-the-art methods by 1-3 dB PSNR.
format Preprint
id arxiv_https___arxiv_org_abs_2603_00492
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Models
de Lutio, Riccardo
Fischer, Tobias
Chang, Yen-Yu
Zhang, Yuxuan
Wu, Jay Zhangjie
Ren, Xuanchi
Shen, Tianchang
Tothova, Katarina
Gojcic, Zan
Turki, Haithem
Computer Vision and Pattern Recognition
Artificial Intelligence
Graphics
Machine Learning
Per-scene optimization methods such as 3D Gaussian Splatting provide state-of-the-art novel view synthesis quality but extrapolate poorly to under-observed areas. Methods that leverage generative priors to correct artifacts in these areas hold promise but currently suffer from two shortcomings. The first is scalability, as existing methods use image diffusion models or bidirectional video models that are limited in the number of views they can generate in a single pass (and thus require a costly iterative distillation process for consistency). The second is quality itself, as generators used in prior work tend to produce outputs that are inconsistent with existing scene content and fail entirely in completely unobserved regions. To solve these, we propose a two-stage pipeline that leverages two key insights. First, we train a powerful bidirectional generative model with a novel opacity mixing strategy that encourages consistency with existing observations while retaining the model's ability to extrapolate novel content in unseen areas. Second, we distill it into a causal auto-regressive model that generates hundreds of frames in a single pass. This model can directly produce novel views or serve as pseudo-supervision to improve the underlying 3D representation in a simple and highly efficient manner. We evaluate our method extensively and demonstrate that it can generate plausible reconstructions in scenarios where existing approaches fail completely. When measured on commonly benchmarked datasets, we outperform all existing baselines by a wide margin, exceeding prior state-of-the-art methods by 1-3 dB PSNR.
title ArtiFixer: Enhancing and Extending 3D Reconstruction with Auto-Regressive Diffusion Models
topic Computer Vision and Pattern Recognition
Artificial Intelligence
Graphics
Machine Learning
url https://arxiv.org/abs/2603.00492