发表机构
CSIRO; Monash University（澳大利亚联邦科学与工业研究组织; 莫纳什大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
针对前馈三维重建在困难条件下静默失败的问题，提出轻量级可靠性适配器GeoCond，利用几何条件预测不确定性并门控细化，显著提升OOD可靠性且支持零样本迁移。
AI 中文摘要
前馈三维基础模型（如VGGT）能够在单次前向传播中预测相机参数、深度和点图，但在低重叠、低视差和极端相对旋转的情况下可能静默失败。对这些因素的层次化分析表明，这些失败由几何条件决定，且原生偶然不确定性难以捕捉。我们提出GeoCond，一种用于冻结的前馈三维骨干网络的轻量级可靠性适配器。GeoCond读取骨干网络预测的几何信息，输出姿态级不确定性和细化门控。在训练时，可通过帧置换轨道方差、有标签时的真值姿态误差或无标签独立姿态图的循环残差进行监督。推理时，默认头仅需一次骨干网络前向传播和一个小型MLP。在VGGT上，GeoCond将分布外（OOD）AUSE（稀疏化误差曲线下面积；越低越好）从原生置信度的0.32提升至0.20，零样本迁移到户外极端视角场景，并避免了统一应用束调整导致的崩溃。在多个骨干网络上，循环蒸馏变体提供了无需真值的适配路径，包括在等变模型上置换方差消失的情况。相同的可靠性信号支持门控细化、姿态图加权、校准、数据筛选和采集决策。可靠的前馈三维重建不仅需要预测几何，还需要知道何时该信任该几何。
英文摘要
Feed-forward 3D foundation models such as VGGT predict cameras, depth, and point maps in a single pass, but can fail silently under low overlap, low parallax, and extreme relative rotation. Stratified analyses over these factors show that these failures are governed by geometric conditioning and are poorly captured by native aleatoric confidence. We introduce GeoCond, a lightweight reliability adapter for frozen feed-forward 3D backbones. GeoCond reads the backbone's predicted geometry and outputs pose-level uncertainty and a refinement gate. During training, it can be supervised by frame-permutation orbit variance, ground-truth pose error when labels are available, or cycle residuals from unlabelled independent pose graphs. At inference, the default head requires only one backbone pass and a small MLP. On VGGT, GeoCond improves out-of-distribution (OOD) AUSE (area under the sparsification-error curve; lower is better) from $0.32$ to $0.20$ over native confidence, transfers zero-shot to outdoor extreme-view scenes, and avoids the collapse caused by applying bundle adjustment uniformly. Across multiple backbones, cycle-distilled variants provide a ground-truth-free adaptation route, including cases where permutation variance vanishes on equivariant models. The same reliability signal supports gated refinement, pose-graph weighting, calibration, curation, and capture decisions. Reliable feed-forward 3D reconstruction requires not only predicting geometry, but also knowing when that geometry should be trusted.
