Back to 3D: Few-Shot 3D Keypoint Detection with Back-Projected 2D Features

TL;DR

This paper introduces a novel few-shot 3D keypoint detection method that leverages back-projected features from large pre-trained 2D vision models, achieving state-of-the-art results on the KeyPointNet dataset.

Contribution

It proposes a new approach combining back-projected 2D features and a keypoint candidate optimization module for improved 3D keypoint detection.

Findings

Achieves near doubling of previous best performance on KeyPointNet dataset.

Demonstrates robustness of 3D features derived from 2D foundation models.

Provides analysis of features from different 2D foundation models.

Abstract

With the immense growth of dataset sizes and computing resources in recent years, so-called foundation models have become popular in NLP and vision tasks. In this work, we propose to explore foundation models for the task of keypoint detection on 3D shapes. A unique characteristic of keypoint detection is that it requires semantic and geometric awareness while demanding high localization accuracy. To address this problem, we propose, first, to back-project features from large pre-trained 2D vision models onto 3D shapes and employ them for this task. We show that we obtain robust 3D features that contain rich semantic information and analyze multiple candidate features stemming from different 2D foundation models. Second, we employ a keypoint candidate optimization module which aims to match the average observed distribution of keypoints on the shape and is guided by the back-projected features. The resulting approach achieves a new state of the art for few-shot keypoint detection on the KeyPointNet dataset, almost doubling the performance of the previous best methods.