Search papers, labs, and topics across Lattice.
This paper introduces a training-free filtering method that leverages per-view Gaussian predictions to enhance the quality of 3D reconstructions by effectively removing transient distractors. By analyzing the inconsistencies in Gaussian representations across multiple input views, the method identifies and eliminates artifacts without the need for retraining or scene-specific adjustments. The approach demonstrates significant improvements in novel-view quality across various reconstruction models and distractor benchmarks while preserving the integrity of clean scene reconstructions.
Transient objects in 3D reconstructions can be filtered out without retraining, significantly enhancing novel-view quality.
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.
