LaS-Comp: Zero-shot 3D Completion with Latent-Spatial Consistency

2026-02-21

AI Summary

This work addresses the challenging problem of zero-shot, category-agnostic 3D shape completion by proposing the first training-free, two-stage approach. The method comprises an explicit replacement stage that preserves the input geometry and an implicit optimization stage that leverages the generative priors of a 3D foundation model to smooth the boundary between observed and completed regions. Notably, this is the first effort to incorporate geometric priors from 3D foundation models into zero-shot completion. To facilitate comprehensive evaluation, the authors introduce Omni-Comp, a new benchmark for category-agnostic 3D shape completion. Extensive experiments demonstrate that the proposed method significantly outperforms current state-of-the-art approaches on Omni-Comp, achieving superior completion quality and generalization capability without requiring any training.

Technology Category

• Computer Vision: 3D Computer Vision
• Machine Learning: Deep Neural Architectures and Foundation Models
• Humans and AI: Other Foundations of Human Computation & AI

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

This paper introduces LaS-Comp, a zero-shot and category-agnostic approach that leverages the rich geometric priors of 3D foundation models to enable 3D shape completion across diverse types of partial observations. Our contributions are threefold: First, \ourname{} harnesses these powerful generative priors for completion through a complementary two-stage design: (i) an explicit replacement stage that preserves the partial observation geometry to ensure faithful completion; and (ii) an implicit refinement stage ensures seamless boundaries between the observed and synthesized regions. Second, our framework is training-free and compatible with different 3D foundation models. Third, we introduce Omni-Comp, a comprehensive benchmark combining real-world and synthetic data with diverse and challenging partial patterns, enabling a more thorough and realistic evaluation. Both quantitative and qualitative experiments demonstrate that our approach outperforms previous state-of-the-art approaches. Our code and data will be available at \href{https://github.com/DavidYan2001/LaS-Comp}{LaS-Comp}.

Problem

• 3D completion
• zero-shot
• partial observations
• category-agnostic
• 3D shape reconstruction

Innovation

• zero-shot
• 3D completion
• foundation models
• training-free
• geometric priors