# Per-View Gaussian Predictions Enable Training-Free Distractor Filtering in Feed-Forward 3DGS

arXiv: https://arxiv.org/abs/2608.26951 · HTML: https://mlsanity.com/paper/2608.26951/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 20 of its 28 references that are in the map.
- First posted on arXiv 27 Aug 2026.

## Builds on (most cited first, 20 of 20)

- [The Unreasonable Effectiveness of Deep Features as a Perceptual Metric](https://mlsanity.com/paper/1801.03924/) (2018)
- [3D Gaussian Splatting for Real-Time Radiance Field Rendering](https://mlsanity.com/paper/2308.04079/) (2023)
- [NeRF in the Wild: Neural Radiance Fields for Unconstrained Photo Collections](https://mlsanity.com/paper/2008.02268/) (2020)
- [4D Gaussian Splatting for Real-Time Dynamic Scene Rendering](https://mlsanity.com/paper/2310.08528/) (2023)
- [Scaffold-GS: Structured 3D Gaussians for View-Adaptive Rendering](https://mlsanity.com/paper/2312.00109/) (2023)
- [pixelSplat: 3D Gaussian Splats from Image Pairs for Scalable Generalizable 3D Reconstruction](https://mlsanity.com/paper/2312.12337/) (2023)
- [PhysGaussian: Physics-Integrated 3D Gaussians for Generative Dynamics](https://mlsanity.com/paper/2311.12198/) (2023)
- [DepthSplat: Connecting Gaussian Splatting and Depth](https://mlsanity.com/paper/2410.13862/) (2024)
- [No Pose, No Problem: Surprisingly Simple 3D Gaussian Splats from Sparse Unposed Images](https://mlsanity.com/paper/2410.24207/) (2024)
- [AnySplat: Feed-forward 3D Gaussian Splatting from Unconstrained Views](https://mlsanity.com/paper/2505.23716/) (2025)
- [CityGaussian: Real-time High-quality Large-Scale Scene Rendering with Gaussians](https://mlsanity.com/paper/2404.01133/) (2024)
- [WildGaussians: 3D Gaussian Splatting in the Wild](https://mlsanity.com/paper/2407.08447/) (2024)
- [RobustNeRF: Ignoring Distractors with Robust Losses](https://mlsanity.com/paper/2302.00833/) (2023)
- [NeRF On-the-go: Exploiting Uncertainty for Distractor-free NeRFs in the Wild](https://mlsanity.com/paper/2405.18715/) (2024)
- [VolSplat: Rethinking Feed-Forward 3D Gaussian Splatting with Voxel-Aligned Prediction](https://mlsanity.com/paper/2509.19297/) (2025)
- [DeSplat: Decomposed Gaussian Splatting for Distractor-Free Rendering](https://mlsanity.com/paper/2411.19756/) (2024)
- [TokenSplat: Token-aligned 3D Gaussian Splatting for Feed-forward Pose-free Reconstruction](https://mlsanity.com/paper/2603.00697/) (2026)
- [ForestSplats: Deformable transient field for Gaussian Splatting in the Wild](https://mlsanity.com/paper/2503.06179/) (2025)
- [DualSplat: Robust 3D Gaussian Splatting via Pseudo-Mask Bootstrapping from Reconstruction Failures](https://mlsanity.com/paper/2604.21631/) (2026)
- [Generalizable Sparse-View 3D Reconstruction from Unconstrained Images](https://mlsanity.com/paper/2604.28193/) (2026)

## Abstract (arXiv)

Feed-forward 3D Gaussian Splatting reconstructs an explicit Gaussian representation from multiple input images in one network execution, making 3D reconstruction increasingly accessible for casual captures. However, such captures frequently contain transient objects that appear in only a subset of the views. Such content can be encoded into the per-view Gaussians associated with the inputs that observe it and remain in the combined representation despite being observed by no other input. As a result, it may produce blurred, duplicated, or floating artifacts in novel views. We introduce a training-free filtering procedure that exploits this per-view prediction structure. For each input, we exclude its associated Gaussians and render the same camera using the remaining representation, revealing content that is inconsistent with the other inputs. Feature similarity forms candidate regions, and rendering-based verification retains only candidates whose removal reduces reconstruction error in the other input views. The procedure operates on a single frozen prediction without retraining or scene-specific optimization. Across three reconstruction models and two distractor benchmarks, it consistently improves novel-view quality with varying numbers of input views. On clean scenes, evaluations across four models show that the original reconstructions are largely preserved.