Neu-PiG: Neural Preconditioned Grids for Fast Dynamic Surface Reconstruction on Long Sequences

2026-02-25

AI Summary

This work addresses the challenging problem of long-term, temporally consistent surface reconstruction of dynamic 3D objects from unstructured point clouds. The authors propose an efficient, drift-free implicit reconstruction method that encodes full temporal deformations by constructing a multi-resolution implicit grid based on position and normal information from a single keyframe, and decodes per-frame 6-DoF deformations via a lightweight MLP. Innovatively, the approach introduces a normal-position parameterized multi-scale implicit representation combined with Sobolev preconditioned gradient optimization, eliminating the need for explicit correspondences or category priors. Experiments demonstrate that the method significantly outperforms existing techniques on human and animal datasets, achieving higher accuracy, scalability to long sequences, and inference speeds over 60 times faster than non-learning-based methods while matching the performance of heavyweight pretrained models.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Learning with Manifolds
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Scalable techniques for the creation, curation, publication, maintenance, and consumption of large, Web-based, structured, reusable, knowledge graphs and ontologies

Abstract

Temporally consistent surface reconstruction of dynamic 3D objects from unstructured point cloud data remains challenging, especially for very long sequences. Existing methods either optimize deformations incrementally, risking drift and requiring long runtimes, or rely on complex learned models that demand category-specific training. We present Neu-PiG, a fast deformation optimization method based on a novel preconditioned latent-grid encoding that distributes spatial features parameterized on the position and normal direction of a keyframe surface. Our method encodes entire deformations across all time steps at various spatial scales into a multi-resolution latent grid, parameterized by the position and normal direction of a reference surface from a single keyframe. This latent representation is then augmented for time modulation and decoded into per-frame 6-DoF deformations via a lightweight multilayer perceptron (MLP). To achieve high-fidelity, drift-free surface reconstructions in seconds, we employ Sobolev preconditioning during gradient-based training of the latent space, completely avoiding the need for any explicit correspondences or further priors. Experiments across diverse human and animal datasets demonstrate that Neu-PiG outperforms state-the-art approaches, offering both superior accuracy and scalability to long sequences while running at least 60x faster than existing training-free methods and achieving inference speeds on the same order as heavy pretrained models.

Problem

• surface reconstruction
• dynamic 3D objects
• point cloud
• temporal consistency
• long sequences

Innovation

• Neural Preconditioning
• Latent Grid Encoding
• Dynamic Surface Reconstruction
• Sobolev Preconditioning
• 6-DoF Deformation