Parameter-efficient Prompt Learning for 3D Point Cloud Understanding
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929256434499584 |
|---|---|
| author | Sun, Hongyu Wang, Yongcai Chen, Wang Deng, Haoran Li, Deying |
| author_facet | Sun, Hongyu Wang, Yongcai Chen, Wang Deng, Haoran Li, Deying |
| contents | This paper presents a parameter-efficient prompt tuning method, named PPT, to adapt a large multi-modal model for 3D point cloud understanding. Existing strategies are quite expensive in computation and storage, and depend on time-consuming prompt engineering. We address the problems from three aspects. Firstly, a PromptLearner module is devised to replace hand-crafted prompts with learnable contexts to automate the prompt tuning process. Then, we lock the pre-trained backbone instead of adopting the full fine-tuning paradigm to substantially improve the parameter efficiency. Finally, a lightweight PointAdapter module is arranged near target tasks to enhance prompt tuning for 3D point cloud understanding. Comprehensive experiments are conducted to demonstrate the superior parameter and data efficiency of the proposed method.Meanwhile, we obtain new records on 4 public datasets and multiple 3D tasks, i.e., point cloud recognition, few-shot learning, and part segmentation. The implementation is available at https://github.com/auniquesun/PPT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_15823 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Parameter-efficient Prompt Learning for 3D Point Cloud Understanding Sun, Hongyu Wang, Yongcai Chen, Wang Deng, Haoran Li, Deying Computer Vision and Pattern Recognition This paper presents a parameter-efficient prompt tuning method, named PPT, to adapt a large multi-modal model for 3D point cloud understanding. Existing strategies are quite expensive in computation and storage, and depend on time-consuming prompt engineering. We address the problems from three aspects. Firstly, a PromptLearner module is devised to replace hand-crafted prompts with learnable contexts to automate the prompt tuning process. Then, we lock the pre-trained backbone instead of adopting the full fine-tuning paradigm to substantially improve the parameter efficiency. Finally, a lightweight PointAdapter module is arranged near target tasks to enhance prompt tuning for 3D point cloud understanding. Comprehensive experiments are conducted to demonstrate the superior parameter and data efficiency of the proposed method.Meanwhile, we obtain new records on 4 public datasets and multiple 3D tasks, i.e., point cloud recognition, few-shot learning, and part segmentation. The implementation is available at https://github.com/auniquesun/PPT. |
| title | Parameter-efficient Prompt Learning for 3D Point Cloud Understanding |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2402.15823 |