PointMAC: Meta-Learned Adaptation for Robust Test-Time Point Cloud Completion

2025-10-11

AI Summary

Point cloud completion for safety-critical applications (e.g., robotics, AR) demands strong robustness, yet existing methods rely on static inference and fixed inductive biases, limiting adaptability to unseen structural patterns and sensor distortions at test time. To address this, we propose the first unsupervised meta-assisted test-time adaptation framework for point cloud completion. Our method extends MAML with dual self-supervised objectives—simulating structural and sensor-level corruptions—to online optimize a shared encoder. Crucially, we introduce an adaptive λ calibration mechanism that dynamically aligns gradient directions between auxiliary and main tasks. Without requiring additional annotations, our approach achieves state-of-the-art performance across synthetic, simulated, and real-world datasets. It significantly improves completion quality under diverse degradation scenarios—including occlusion, noise, and resolution mismatch—demonstrating superior generalization and stability.

Technology Category

• Machine Learning: Calibration & Uncertainty Quantification
• Computer Vision: Adversarial Attacks & Robustness
• Search and Optimization: Metareasoning and Metaheuristics

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Systems and Infrastructure for Web, Mobile and WoT: Applied ML and AI for Web-based mobile applications

Abstract

Point cloud completion is essential for robust 3D perception in safety-critical applications such as robotics and augmented reality. However, existing models perform static inference and rely heavily on inductive biases learned during training, limiting their ability to adapt to novel structural patterns and sensor-induced distortions at test time. To address this limitation, we propose PointMAC, a meta-learned framework for robust test-time adaptation in point cloud completion. It enables sample-specific refinement without requiring additional supervision. Our method optimizes the completion model under two self-supervised auxiliary objectives that simulate structural and sensor-level incompleteness. A meta-auxiliary learning strategy based on Model-Agnostic Meta-Learning (MAML) ensures that adaptation driven by auxiliary objectives is consistently aligned with the primary completion task. During inference, we adapt the shared encoder on-the-fly by optimizing auxiliary losses, with the decoder kept fixed. To further stabilize adaptation, we introduce Adaptive $λ$-Calibration, a meta-learned mechanism for balancing gradients between primary and auxiliary objectives. Extensive experiments on synthetic, simulated, and real-world datasets demonstrate that PointMAC achieves state-of-the-art results by refining each sample individually to produce high-quality completions. To the best of our knowledge, this is the first work to apply meta-auxiliary test-time adaptation to point cloud completion.

Problem

• Adapting point cloud completion to novel structural patterns
• Handling sensor-induced distortions during test-time inference
• Enabling sample-specific refinement without additional supervision

Innovation

• Meta-learned adaptation for test-time point cloud completion
• Self-supervised auxiliary objectives simulate structural incompleteness
• Meta-learned gradient balancing between primary and auxiliary objectives