Abstract: A key challenge in reconstructing high-quality 3D scans is registering data from different viewpoints. Existing global (multiview) alignment algorithms are restricted to rigid-body transformations, and cannot adequately handle non-rigid warps frequently present in real-world datasets. Moreover, algorithms that can compensate for such warps between pairs of scans do not easily generalize to the multiview case. We present an algorithm for obtaining a globally optimal alignment of multiple overlapping datasets in the presence of low-frequency non-rigid deformations, such as those caused by device nonlinearities or calibration error. The process first obtains sparse correspondences between views using a locally weighted, stability-guaranteeing variant of iterative closest points (ICP). Global positions for feature points are found using a relaxation method, and the scans are warped to their final positions using thin-plate splines. Our framework efficiently handles large datasets---thousands of scans comprising hundreds of millions of samples---for both rigid and non-rigid alignment, with the non-rigid case requiring little overhead beyond rigid-body alignment. We demonstrate that, relative to rigid-body registration, it improves the quality of alignment and better preserves detail in 3D datasets from a variety of scanners exhibiting non-rigid distortion.