Make Your ViT-based Multi-view 3D Detectors Faster via Token Compression
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_ | 1866912010601496576 |
|---|---|
| author | Zhang, Dingyuan Liang, Dingkang Tan, Zichang Ye, Xiaoqing Zhang, Cheng Wang, Jingdong Bai, Xiang |
| author_facet | Zhang, Dingyuan Liang, Dingkang Tan, Zichang Ye, Xiaoqing Zhang, Cheng Wang, Jingdong Bai, Xiang |
| contents | Slow inference speed is one of the most crucial concerns for deploying multi-view 3D detectors to tasks with high real-time requirements like autonomous driving. Although many sparse query-based methods have already attempted to improve the efficiency of 3D detectors, they neglect to consider the backbone, especially when using Vision Transformers (ViT) for better performance. To tackle this problem, we explore the efficient ViT backbones for multi-view 3D detection via token compression and propose a simple yet effective method called TokenCompression3D (ToC3D). By leveraging history object queries as foreground priors of high quality, modeling 3D motion information in them, and interacting them with image tokens through the attention mechanism, ToC3D can effectively determine the magnitude of information densities of image tokens and segment the salient foreground tokens. With the introduced dynamic router design, ToC3D can weigh more computing resources to important foreground tokens while compressing the information loss, leading to a more efficient ViT-based multi-view 3D detector. Extensive results on the large-scale nuScenes dataset show that our method can nearly maintain the performance of recent SOTA with up to 30% inference speedup, and the improvements are consistent after scaling up the ViT and input resolution. The code will be made at https://github.com/DYZhang09/ToC3D. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_00633 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Make Your ViT-based Multi-view 3D Detectors Faster via Token Compression Zhang, Dingyuan Liang, Dingkang Tan, Zichang Ye, Xiaoqing Zhang, Cheng Wang, Jingdong Bai, Xiang Computer Vision and Pattern Recognition Slow inference speed is one of the most crucial concerns for deploying multi-view 3D detectors to tasks with high real-time requirements like autonomous driving. Although many sparse query-based methods have already attempted to improve the efficiency of 3D detectors, they neglect to consider the backbone, especially when using Vision Transformers (ViT) for better performance. To tackle this problem, we explore the efficient ViT backbones for multi-view 3D detection via token compression and propose a simple yet effective method called TokenCompression3D (ToC3D). By leveraging history object queries as foreground priors of high quality, modeling 3D motion information in them, and interacting them with image tokens through the attention mechanism, ToC3D can effectively determine the magnitude of information densities of image tokens and segment the salient foreground tokens. With the introduced dynamic router design, ToC3D can weigh more computing resources to important foreground tokens while compressing the information loss, leading to a more efficient ViT-based multi-view 3D detector. Extensive results on the large-scale nuScenes dataset show that our method can nearly maintain the performance of recent SOTA with up to 30% inference speedup, and the improvements are consistent after scaling up the ViT and input resolution. The code will be made at https://github.com/DYZhang09/ToC3D. |
| title | Make Your ViT-based Multi-view 3D Detectors Faster via Token Compression |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.00633 |