Exploiting Vision Language Model for Training-Free 3D Point Cloud OOD Detection via Graph Score Propagation

TL;DR

This paper presents a training-free approach using Vision-Language Models and graph-based score propagation for improved out-of-distribution detection in 3D point clouds, addressing a critical challenge in robust perception.

Contribution

It introduces a novel Graph Score Propagation method that leverages VLMs for 3D OOD detection without training, incorporating prompt clustering and self-training for enhanced performance.

Findings

GSP outperforms existing methods on synthetic datasets

Effective in few-shot scenarios

Works well on real-world 3D point cloud data

Abstract

Out-of-distribution (OOD) detection in 3D point cloud data remains a challenge, particularly in applications where safe and robust perception is critical. While existing OOD detection methods have shown progress for 2D image data, extending these to 3D environments involves unique obstacles. This paper introduces a training-free framework that leverages Vision-Language Models (VLMs) for effective OOD detection in 3D point clouds. By constructing a graph based on class prototypes and testing data, we exploit the data manifold structure to enhancing the effectiveness of VLMs for 3D OOD detection. We propose a novel Graph Score Propagation (GSP) method that incorporates prompt clustering and self-training negative prompting to improve OOD scoring with VLM. Our method is also adaptable to few-shot scenarios, providing options for practical applications. We demonstrate that GSP consistently outperforms state-of-the-art methods across synthetic and real-world datasets 3D point cloud OOD detection.