ReFu: Recursive Fusion for Exemplar-Free 3D Class-Incremental Learning
Fuente:
arXiv
Saved in:
| Main Authors: | , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915177458302976 |
|---|---|
| author | Yang, Yi Zhong, Lei Zhuang, Huiping |
| author_facet | Yang, Yi Zhong, Lei Zhuang, Huiping |
| contents | We introduce a novel Recursive Fusion model, dubbed ReFu, designed to integrate point clouds and meshes for exemplar-free 3D Class-Incremental Learning, where the model learns new 3D classes while retaining knowledge of previously learned ones. Unlike existing methods that either rely on storing historical data to mitigate forgetting or focus on single data modalities, ReFu eliminates the need for exemplar storage while utilizing the complementary strengths of both point clouds and meshes. To achieve this, we introduce a recursive method which continuously accumulates knowledge by updating the regularized auto-correlation matrix. Furthermore, we propose a fusion module, featuring a Pointcloud-guided Mesh Attention Layer that learns correlations between the two modalities. This mechanism effectively integrates point cloud and mesh features, leading to more robust and stable continual learning. Experiments across various datasets demonstrate that our proposed framework outperforms existing methods in 3D class-incremental learning. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_12326 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | ReFu: Recursive Fusion for Exemplar-Free 3D Class-Incremental Learning Yang, Yi Zhong, Lei Zhuang, Huiping Computer Vision and Pattern Recognition We introduce a novel Recursive Fusion model, dubbed ReFu, designed to integrate point clouds and meshes for exemplar-free 3D Class-Incremental Learning, where the model learns new 3D classes while retaining knowledge of previously learned ones. Unlike existing methods that either rely on storing historical data to mitigate forgetting or focus on single data modalities, ReFu eliminates the need for exemplar storage while utilizing the complementary strengths of both point clouds and meshes. To achieve this, we introduce a recursive method which continuously accumulates knowledge by updating the regularized auto-correlation matrix. Furthermore, we propose a fusion module, featuring a Pointcloud-guided Mesh Attention Layer that learns correlations between the two modalities. This mechanism effectively integrates point cloud and mesh features, leading to more robust and stable continual learning. Experiments across various datasets demonstrate that our proposed framework outperforms existing methods in 3D class-incremental learning. |
| title | ReFu: Recursive Fusion for Exemplar-Free 3D Class-Incremental Learning |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.12326 |