FixAnything: 3D-Consistent Rendering Refinement via Video Generative Priors
2026-08-24 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors present FixAnything, a method to fix visual mistakes in 3D rendered images made from different 3D scene types like Gaussian Splatting and Neural Radiance Fields. Instead of building separate fixes for each type, they use a pre-trained video generation model to clean up the images by treating the problem as a video-to-video translation task. They guide the fix using masks that show which parts of the image are already good and use camera pose data to help keep the 3D structure consistent. Their approach works across multiple 3D representations with minimal fine-tuning and doesn’t need custom designs for each case.
3D scene representationsGaussian SplattingNeural Radiance Fields (NeRF)rendering artifactsdiffusion priorsvideo generative modelvideo-to-video translationcamera posestructure-from-motionfine-tuning
Authors
Khiem Vuong, Deva Ramanan, Srinivasa Narasimhan
Abstract
Rendering views using 3D scene representations such as Gaussian Splatting (3DGS), Neural Radiance Fields (NeRF), meshes, or even point clouds produces artifacts when input views are sparse or target views lie far from the input. Recent work mitigates these artifacts using diffusion-based generative priors, but is specialized to individual representations and require custom architectures or extensive retraining. We present FixAnything, a single model for fixing a wide range of rendering artifacts. It does so by repurposing a pretrained video generative model, leveraging its implicit multi-view priors with only minimal modification and lightweight finetuning. Our key insight is that even noisily-rendered sequences preserve camera motion and coarse scene structure, allowing cleanup to be formulated as video-to-video translation. To control what scene structure should be preserved, we introduce a binary mask denoting the clean pixels, enabling the model to anchor its output to high-quality inputs (e.g. training views) while refining the rest. To encourage FixAnything to produce 3D-consistent renderings that support downstream reconstruction, we use camera pose accuracy (recovered via structure-from-motion) as a reward signal for direct preference optimization (DPO). Across four distinct 3D representations, FixAnything consistently improves rendering quality with lightweight finetuning, demonstrating that a single generalist video prior can replace multiple specialist refinement pipelines. The simplicity of the framework enables immediate adoption of stronger future video models without architectural redesign.