IGFuse: Interactive 3D Gaussian Scene Reconstruction via Multi-Scans Fusion

TL;DR

IGFuse is a novel framework that fuses multiple 3D scans using Gaussian fields to reconstruct complete, interactive scenes with high fidelity, overcoming occlusions and sensor limitations without complex pipelines.

Contribution

It introduces a segmentation-aware Gaussian fusion method with bi-directional consistency and a pseudo-intermediate scene for robust alignment, enabling scalable, high-quality 3D scene reconstruction.

Findings

Effective in reconstructing complete scenes with occlusions

Generalizes well to unseen scene configurations

Enables high-fidelity rendering and object manipulation

Abstract

Reconstructing complete and interactive 3D scenes remains a fundamental challenge in computer vision and robotics, particularly due to persistent object occlusions and limited sensor coverage. Multiview observations from a single scene scan often fail to capture the full structural details. Existing approaches typically rely on multi stage pipelines, such as segmentation, background completion, and inpainting or require per-object dense scanning, both of which are error-prone, and not easily scalable. We propose IGFuse, a novel framework that reconstructs interactive Gaussian scene by fusing observations from multiple scans, where natural object rearrangement between captures reveal previously occluded regions. Our method constructs segmentation aware Gaussian fields and enforces bi-directional photometric and semantic consistency across scans. To handle spatial misalignments, we introduce a pseudo-intermediate scene state for unified alignment, alongside collaborative co-pruning strategies to refine geometry. IGFuse enables high fidelity rendering and object level scene manipulation without dense observations or complex pipelines. Extensive experiments validate the framework's strong generalization to novel scene configurations, demonstrating its effectiveness for real world 3D reconstruction and real-to-simulation transfer. Our project page is available online.