GenPC: Zero-shot Point Cloud Completion via 3D Generative Priors

TL;DR

GenPC introduces a zero-shot point cloud completion method that leverages 3D generative priors trained on large-scale data, enabling high-quality reconstruction of real-world scans without relying on synthetic training datasets.

Contribution

The paper proposes a novel zero-shot completion framework using 3D generative models and introduces modules to connect partial scans with generative priors while preserving original structures.

Findings

Outperforms existing methods on standard benchmarks.

Demonstrates strong generalization to real-world scans.

Effectively preserves partial structure during completion.

Abstract

Existing point cloud completion methods, which typically depend on predefined synthetic training datasets, encounter significant challenges when applied to out-of-distribution, real-world scans. To overcome this limitation, we introduce a zero-shot completion framework, termed GenPC, designed to reconstruct high-quality real-world scans by leveraging explicit 3D generative priors. Our key insight is that recent feed-forward 3D generative models, trained on extensive internet-scale data, have demonstrated the ability to perform 3D generation from single-view images in a zero-shot setting. To harness this for completion, we first develop a Depth Prompting module that links partial point clouds with image-to-3D generative models by leveraging depth images as a stepping stone. To retain the original partial structure in the final results, we design the Geometric Preserving Fusion module that aligns the generated shape with input by adaptively adjusting its pose and scale. Extensive experiments on widely used benchmarks validate the superiority and generalizability of our approach, bringing us a step closer to robust real-world scan completion.