发表机构
University of Waterloo; Aalto University（滑铁卢大学; 阿尔托大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究通过控制稀疏输入的视角跨度，系统测试前馈式三维重建模型，发现宽视角导致表面覆盖不全和几何失真，揭示其分布偏移下的脆弱性。
AI 中文摘要
前馈式三维重建模型能够从稀疏图像中高效估计几何结构，但其预训练特性可能使其在面对训练数据之外的分布偏移时变得脆弱。识别这些失败模式对于理解此类模型在无约束成像场景中能否可靠部署至关重要。我们通过保持输入预算固定、改变稀疏图像输入的视角跨度，将视角变化作为一种受控的分布偏移进行研究。在多个前馈式重建模型上，我们观察到随着视角跨度的增加，性能出现显著下降，宽视角跨度既导致表面覆盖不完整，也产生观测图像不支持的几何结构。这些结果表明，视角变化可能引发超出常规重建不完整性的失败模式，凸显了在挑战预训练模型所学几何先验的分布偏移下评估前馈式模型的必要性。
英文摘要
Feed-forward 3D reconstruction models enable efficient geometry estimation from sparse images, but their pretrained nature can make them vulnerable to distribution shifts beyond their training data. Identifying these failure modes is important for understanding when such models can be reliably deployed in unconstrained imaging settings. We investigate viewpoint variation as a controlled distribution shift by varying the angular span of sparse image inputs while keeping the input budget fixed. Across multiple feed-forward reconstruction models, we observe substantial degradation as viewpoint span increases, with wide spans producing both incomplete surface coverage and geometry unsupported by the observed imagery. These results reveal that viewpoint variation can induce failure modes beyond conventional reconstruction incompleteness, highlighting the need to evaluate pretrained feed-forward models under distribution shifts that challenge their learned geometric priors.
