# Direct Object-Level Reconstruction via Probabilistic Gaussian Splatting

Source: [Direct Object-Level Reconstruction via Probabilistic Gaussian Splatting](<https://www.sota2.com/research/paper/69b8e03d0636e5bf4afbed97>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-07-15T05:00:09.989Z

Authors: Shuai Guo, Ao Guo, Junchao Zhao, Qi Chen, Yuxiang Qi, Zechuan Li, Dong Chen, Mingliang Xu

Published: 2026-03-15T10:17:14.000Z

## Abstract

Object-level 3D reconstruction play important roles across domains such as cultural heritage digitization, industrial manufacturing, and virtual reality. However, existing Gaussian Splatting-based approaches generally rely on full-scene reconstruction, in which substantial redundant background information is introduced, leading to increased computational and storage overhead. To address this limitation, we propose an efficient single-object 3D reconstruction method based on 2D Gaussian Splatting. By directly integrating foreground-background probability cues into Gaussian primitives and dynamically pruning low-probability Gaussians during training, the proposed method fundamentally focuses on an object of interest and improves the memory and computational efficiency. Our pipeline leverages probability masks generated by YOLO and SAM to supervise probabilistic Gaussian attributes, replacing binary masks with continuous probability values to mitigate boundary ambiguity. Additionally, we propose a dual-stage filtering strategy for training's startup to suppress background Gaussians. And, during training, rendered probability masks are conversely employed to refine supervision and enhance boundary consistency across views. Experiments conducted on the MIP-360, T&T, and NVOS datasets demonstrate that our method exhibits strong self-correction capability in the presence of mask errors and achieves reconstruction quality comparable to standard 3DGS approaches, while requiring only approximately 1/10 of their Gaussian amount. These results validate the efficiency and robustness of our method for single-object reconstruction and highlight its potential for applications requiring both high fidelity and computational efficiency.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Single-object Reconstruction on NVOS](<https://www.sota2.com/research/sota/single-object-reconstruction-on-nvos.md>): Single-object Reconstruction; NVOS; mIOU: 90.1.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.14316>)

[PDF](<https://arxiv.org/pdf/2603.14316.pdf>)
