Positional Prompt Tuning for Efficient 3D Representation Learning

TL;DR

This paper introduces PPT, a parameter-efficient fine-tuning method for 3D point cloud analysis that leverages positional prompts and achieves state-of-the-art results with minimal trainable parameters.

Contribution

The paper proposes PPT, a novel PEFT approach for 3D representation learning that uses increased patch tokens and trainable positional encoding, while keeping most pre-trained model parameters frozen.

Findings

PPT achieves 95.01% accuracy on ScanObjectNN OBJ_BG dataset.

PPT requires only 1.05 million trainable parameters.

Extensive experiments validate PPT's effectiveness and efficiency.

Abstract

We rethink the role of positional encoding in 3D representation learning and fine-tuning. We argue that using positional encoding in point Transformer-based methods serves to aggregate multi-scale features of point clouds. Additionally, we explore parameter-efficient fine-tuning (PEFT) through the lens of prompts and adapters, introducing a straightforward yet effective method called PPT for point cloud analysis. PPT incorporates increased patch tokens and trainable positional encoding while keeping most pre-trained model parameters frozen. Extensive experiments validate that PPT is both effective and efficient. Our proposed method of PEFT tasks, namely PPT, with only 1.05M of parameters for training, gets state-of-the-art results in several mainstream datasets, such as 95.01% accuracy in the ScanObjectNN OBJ_BG dataset. Codes and weights will be released at https://github.com/zsc000722/PPT.