Training-Free Point Cloud Recognition Based on Geometric and Semantic Information Fusion

TL;DR

This paper introduces a novel training-free point cloud recognition method that fuses geometric and semantic features, reducing computational costs while improving accuracy on benchmark datasets.

Contribution

It is the first to combine geometric and semantic features in a training-free framework for point cloud recognition.

Findings

Outperforms existing training-free methods on ModelNet and ScanObjectNN datasets.

Uses a non-parametric approach for geometric feature extraction.

Incorporates modules to enhance performance in few-shot scenarios.

Abstract

The trend of employing training-free methods for point cloud recognition is becoming increasingly popular due to its significant reduction in computational resources and time costs. However, existing approaches are limited as they typically extract either geometric or semantic features. To address this limitation, we are the first to propose a novel training-free method that integrates both geometric and semantic features. For the geometric branch, we adopt a non-parametric strategy to extract geometric features. In the semantic branch, we leverage a model aligned with text features to obtain semantic features. Additionally, we introduce the GFE module to complement the geometric information of point clouds and the MFF module to improve performance in few-shot settings. Experimental results demonstrate that our method outperforms existing state-of-the-art training-free approaches on mainstream benchmark datasets, including ModelNet and ScanObiectNN.