在野外環境中利用非對稱雙重3D高斯潑濺實現穩健神經渲染

Robust Neural Rendering in the Wild with Asymmetric Dual 3D Gaussian Splatting

June 4, 2025

作者: Chengqi Li, Zhihao Shi, Yangdi Lu, Wenbo He, Xiangyu Xu

cs.AI

摘要

從野外圖像進行3D重建仍然是一項具有挑戰性的任務，這主要歸因於不一致的照明條件和瞬態干擾物。現有方法通常依賴於啟發式策略來處理低質量的訓練數據，這些策略往往難以產生穩定且一致的重建結果，經常導致視覺偽影。在本研究中，我們提出了非對稱雙3DGS（Asymmetric Dual 3DGS），這是一種新穎的框架，它利用了這些偽影的隨機性特點：由於微小的隨機性，它們在不同的訓練運行中往往會有所不同。具體而言，我們的方法並行訓練兩個3D高斯潑濺（3DGS）模型，並施加一致性約束，以促進在可靠場景幾何上的收斂，同時抑制不一致的偽影。為了防止兩個模型由於確認偏誤而陷入相似的失敗模式，我們引入了一種分離掩碼策略，該策略應用兩種互補的掩碼：多線索自適應掩碼和自監督軟掩碼，這導致了兩個模型的非對稱訓練過程，減少了共享的錯誤模式。此外，為了提高模型訓練的效率，我們引入了一種輕量級變體，稱為動態EMA代理（Dynamic EMA Proxy），它用動態更新的指數移動平均（EMA）代理替換了兩個模型中的一個，並採用交替掩碼策略來保持分離。在具有挑戰性的真實世界數據集上的廣泛實驗表明，我們的方法在實現高效率的同時，始終優於現有方法。代碼和訓練好的模型將會發布。

English

3D reconstruction from in-the-wild images remains a challenging task due to inconsistent lighting conditions and transient distractors. Existing methods typically rely on heuristic strategies to handle the low-quality training data, which often struggle to produce stable and consistent reconstructions, frequently resulting in visual artifacts. In this work, we propose Asymmetric Dual 3DGS, a novel framework that leverages the stochastic nature of these artifacts: they tend to vary across different training runs due to minor randomness. Specifically, our method trains two 3D Gaussian Splatting (3DGS) models in parallel, enforcing a consistency constraint that encourages convergence on reliable scene geometry while suppressing inconsistent artifacts. To prevent the two models from collapsing into similar failure modes due to confirmation bias, we introduce a divergent masking strategy that applies two complementary masks: a multi-cue adaptive mask and a self-supervised soft mask, which leads to an asymmetric training process of the two models, reducing shared error modes. In addition, to improve the efficiency of model training, we introduce a lightweight variant called Dynamic EMA Proxy, which replaces one of the two models with a dynamically updated Exponential Moving Average (EMA) proxy, and employs an alternating masking strategy to preserve divergence. Extensive experiments on challenging real-world datasets demonstrate that our method consistently outperforms existing approaches while achieving high efficiency. Codes and trained models will be released.