P2P: Tuning Pre-trained Image Models for Point Cloud Analysis with Point-to-Pixel Prompting

4 Aug 2022  ·  Ziyi Wang, Xumin Yu, Yongming Rao, Jie zhou, Jiwen Lu ·

Nowadays, pre-training big models on large-scale datasets has become a crucial topic in deep learning. The pre-trained models with high representation ability and transferability achieve a great success and dominate many downstream tasks in natural language processing and 2D vision. However, it is non-trivial to promote such a pretraining-tuning paradigm to the 3D vision, given the limited training data that are relatively inconvenient to collect. In this paper, we provide a new perspective of leveraging pre-trained 2D knowledge in 3D domain to tackle this problem, tuning pre-trained image models with the novel Point-to-Pixel prompting for point cloud analysis at a minor parameter cost. Following the principle of prompting engineering, we transform point clouds into colorful images with geometry-preserved projection and geometry-aware coloring to adapt to pre-trained image models, whose weights are kept frozen during the end-to-end optimization of point cloud analysis tasks. We conduct extensive experiments to demonstrate that cooperating with our proposed Point-to-Pixel Prompting, better pre-trained image model will lead to consistently better performance in 3D vision. Enjoying prosperous development from image pre-training field, our method attains 89.3% accuracy on the hardest setting of ScanObjectNN, surpassing conventional point cloud models with much fewer trainable parameters. Our framework also exhibits very competitive performance on ModelNet classification and ShapeNet Part Segmentation. Code is available at https://github.com/wangzy22/P2P.

PDF Abstract

Results from the Paper


Ranked #16 on 3D Point Cloud Classification on ScanObjectNN (using extra training data)

     Get a GitHub badge
Task Dataset Model Metric Name Metric Value Global Rank Uses Extra
Training Data
Result Benchmark
3D Point Cloud Classification ModelNet40 P2P Overall Accuracy 94.0 # 21
Mean Accuracy 91.6 # 8
3D Point Cloud Classification ScanObjectNN P2P Overall Accuracy 89.3 # 16
Number of params 195.8M # 60
3D Part Segmentation ShapeNet-Part P2P Instance Average IoU 86.5 # 18

Methods


No methods listed for this paper. Add relevant methods here