GeoFormer: Learning Point Cloud Completion with Tri-Plane Integrated Transformer
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_ | 1866909286096961536 |
|---|---|
| author | Yu, Jinpeng Huang, Binbin Zhang, Yuxuan Li, Huaxia Tang, Xu Gao, Shenghua |
| author_facet | Yu, Jinpeng Huang, Binbin Zhang, Yuxuan Li, Huaxia Tang, Xu Gao, Shenghua |
| contents | Point cloud completion aims to recover accurate global geometry and preserve fine-grained local details from partial point clouds. Conventional methods typically predict unseen points directly from 3D point cloud coordinates or use self-projected multi-view depth maps to ease this task. However, these gray-scale depth maps cannot reach multi-view consistency, consequently restricting the performance. In this paper, we introduce a GeoFormer that simultaneously enhances the global geometric structure of the points and improves the local details. Specifically, we design a CCM Feature Enhanced Point Generator to integrate image features from multi-view consistent canonical coordinate maps (CCMs) and align them with pure point features, thereby enhancing the global geometry feature. Additionally, we employ the Multi-scale Geometry-aware Upsampler module to progressively enhance local details. This is achieved through cross attention between the multi-scale features extracted from the partial input and the features derived from previously estimated points. Extensive experiments on the PCN, ShapeNet-55/34, and KITTI benchmarks demonstrate that our GeoFormer outperforms recent methods, achieving the state-of-the-art performance. Our code is available at \href{https://github.com/Jinpeng-Yu/GeoFormer}{https://github.com/Jinpeng-Yu/GeoFormer}. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2408_06596 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | GeoFormer: Learning Point Cloud Completion with Tri-Plane Integrated Transformer Yu, Jinpeng Huang, Binbin Zhang, Yuxuan Li, Huaxia Tang, Xu Gao, Shenghua Computer Vision and Pattern Recognition Point cloud completion aims to recover accurate global geometry and preserve fine-grained local details from partial point clouds. Conventional methods typically predict unseen points directly from 3D point cloud coordinates or use self-projected multi-view depth maps to ease this task. However, these gray-scale depth maps cannot reach multi-view consistency, consequently restricting the performance. In this paper, we introduce a GeoFormer that simultaneously enhances the global geometric structure of the points and improves the local details. Specifically, we design a CCM Feature Enhanced Point Generator to integrate image features from multi-view consistent canonical coordinate maps (CCMs) and align them with pure point features, thereby enhancing the global geometry feature. Additionally, we employ the Multi-scale Geometry-aware Upsampler module to progressively enhance local details. This is achieved through cross attention between the multi-scale features extracted from the partial input and the features derived from previously estimated points. Extensive experiments on the PCN, ShapeNet-55/34, and KITTI benchmarks demonstrate that our GeoFormer outperforms recent methods, achieving the state-of-the-art performance. Our code is available at \href{https://github.com/Jinpeng-Yu/GeoFormer}{https://github.com/Jinpeng-Yu/GeoFormer}. |
| title | GeoFormer: Learning Point Cloud Completion with Tri-Plane Integrated Transformer |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2408.06596 |