CanFields: Consolidating Diffeomorphic Flows for Non-Rigid 4D Interpolation from Arbitrary-Length Sequences

TL;DR

CanFields is a new method that interpolates arbitrary-length 3D point cloud sequences into a continuous, deforming shape by jointly optimizing geometry and motion with novel modules, outperforming prior methods especially with noisy or incomplete data.

Contribution

The paper introduces CanFields, a novel approach that consolidates diffeomorphic flows for non-rigid 4D interpolation, featuring a dynamic consolidator and a diffeomorphic motion representation, advancing robustness and detail preservation.

Findings

Validated on over 50 diverse sequences

Superior performance with noisy and incomplete data

Joint optimization of geometry and motion

Abstract

We introduce Canonical Consolidation Fields (CanFields). This novel method interpolates arbitrary-length sequences of independently sampled 3D point clouds into a unified, continuous, and coherent deforming shape. Unlike prior methods that oversmooth geometry or produce topological and geometric artifacts, CanFields optimizes fine-detailed geometry and deformation jointly in an unsupervised fitting with two novel bespoke modules. First, we introduce a dynamic consolidator module that adjusts the input and assigns confidence scores, balancing the optimization of the canonical shape and its motion. Second, we represent the motion as a diffeomorphic flow parameterized by a smooth velocity field. We have validated our robustness and accuracy on more than 50 diverse sequences, demonstrating its superior performance even with missing regions, noisy raw scans, and sparse data. Our project page is at: https://wangmiaowei.github.io/CanFields.github.io/.