Test Time Adaptation Methods for Point Cloud Registration in Laparoscopic Surgery

2026-08-03

AI Summary

This work addresses the domain shift between preoperative synthetic point clouds and intraoperative real point clouds in laparoscopic surgery, caused by noise, sparsity, and occlusion. It is the first to systematically introduce test-time adaptation (TTA) into 3D point cloud registration and tailor it for asymmetric input scenarios. Departing from conventional paradigms that rely on classification logits or prototypes, the proposed method optimizes correspondences directly, integrating auxiliary task–driven model updates, backpropagation-free feature purification and alignment, and layer normalization calibration. Experiments demonstrate that the approach significantly improves registration accuracy on the P2P and P2ILReg datasets, with input adaptation achieving the most consistent error reduction on real-world data while incurring the lowest computational overhead.

Abstract

3D point cloud registration in laparoscopic surgery estimates the transformation between an intraoperative organ reconstructed from video and its preoperative mesh. Because ground-truth transformations are unavailable for real data, supervised networks are trained on synthetic organ pairs. At test time, real reconstructions differ from synthetic data and are noisy, sparse, and occluded, which degrades correspondence estimation. Test-time adaptation (TTA) can reduce this domain shift, but existing methods mainly rely on logits, entropy, class prototypes, or cache memories unavailable in registration. Registration also involves paired inputs with an asymmetric shift that primarily affects the intraoperative cloud. We analyse and modify state-of-the-art TTA methods from three families to 3D registration: model, normalization, and input adaptation. We analyze four representative approaches based on auxiliary-task model updates, backpropagation-free token purging, feature alignment, and layer-normalization calibration. We modify them to handle asymmetric shifts between preoperative and intraoperative point clouds and replace classification-based entropy objectives. Using a correspondence-based model trained on clean synthetic source data, we evaluate adaptation to corrupted synthetic and real target data on P2P and P2ILReg. For synthetic targets, we apply eight corruptions, including uniform noise and global density reduction, at five severity levels. All methods improve registration on P2P, whereas normalization adaptation degrades performance on P2ILReg. Considering the computational overhead of backpropagation-based adaptation, input adaptation is the most promising option for laparoscopic surgery, providing low inference latency and consistent error reductions across datasets. Code: https://github.com/ninaa-git/survey_pc_registration_tta

Problem

• point cloud registration
• test-time adaptation
• domain shift
• laparoscopic surgery
• 3D correspondence

Innovation

• Test-Time Adaptation
• Point Cloud Registration
• Domain Shift
• Asymmetric Adaptation
• Laparoscopic Surgery