Feed-forward 3D reconstruction models are efficient but rigid: once trained, they perform inference in a zero-shot manner and cannot adapt to the test scene. As a result, visually plausible reconstructions often contain errors, particularly under occlusions, specularities, and ambiguous cues. To address this, we introduce Free Geometry, a framework that enables feed-forward 3D reconstruction models to self-evolve at test time without any 3D ground truth. Our key insight is that, when the model receives more views, it produces more reliable and view-consistent reconstructions. Leveraging this property, given a testing sequence, we mask a subset of frames to construct a self-supervised task. Free Geometry enforces cross-view feature consistency between representations from full and partial observations, while maintaining the pairwise relations implied by the held-out frames. This self-supervision allows for fast recalibration via lightweight LoRA updates, taking less than 2 minutes per dataset on a single GPU. Our approach consistently improves state-of-the-art foundation models, including Depth Anything 3 and VGGT, across 4 benchmark datasets, yielding an average improvement of 3.73% in camera pose accuracy and 2.88% in point map prediction. Code is available at https://github.com/hiteacherIamhumble/Free-Geometry .

翻译：前馈式3D重建模型高效但僵化：一旦训练完成，便以零样本方式执行推理，无法适应测试场景。因此，视觉上合理的重建常包含误差，尤其在遮挡、镜面反射和歧义线索条件下。为解决此问题，我们提出自由几何（Free Geometry）框架，该框架使前馈式3D重建模型能在不依赖任何3D真值的情况下，于测试阶段实现自我演化。我们的核心洞察在于：当模型获得更多视角时，其生成的重建结果更可靠且具有视角一致性。利用这一特性，给定测试序列，我们掩膜部分帧以构建自监督任务。自由几何强制完整观测与部分观测表征之间的跨视角特征一致性，同时保持由保留帧隐含的成对关系。这种自监督机制通过轻量级LoRA更新实现快速重校准，在单GPU上每个数据集耗时不足2分钟。我们的方法在4个基准数据集上持续改进包括Depth Anything 3和VGGT在内的最新基础模型，相机位姿精度平均提升3.73%，点图预测精度平均提升2.88%。代码开源于https://github.com/hiteacherIamhumble/Free-Geometry。