Distilling Diffusion Models to Efficient 3D LiDAR Scene Completion
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_ | 1866915412026851328 |
|---|---|
| author | Zhang, Shengyuan Zhao, An Yang, Ling Li, Zejian Meng, Chenye Xu, Haoran Chen, Tianrun Wei, AnYang GU, Perry Pengyun Sun, Lingyun |
| author_facet | Zhang, Shengyuan Zhao, An Yang, Ling Li, Zejian Meng, Chenye Xu, Haoran Chen, Tianrun Wei, AnYang GU, Perry Pengyun Sun, Lingyun |
| contents | Diffusion models have been applied to 3D LiDAR scene completion due to their strong training stability and high completion quality. However, the slow sampling speed limits the practical application of diffusion-based scene completion models since autonomous vehicles require an efficient perception of surrounding environments. This paper proposes a novel distillation method tailored for 3D Li- DAR scene completion models, dubbed ScoreLiDAR, which achieves efficient yet high-quality scene completion. Score- LiDAR enables the distilled model to sample in significantly fewer steps after distillation. To improve completion quality, we also introduce a novel Structural Loss, which encourages the distilled model to capture the geometric structure of the 3D LiDAR scene. The loss contains a scene-wise term constraining the holistic structure and a point-wise term constraining the key landmark points and their relative configuration. Extensive experiments demonstrate that ScoreLiDAR significantly accelerates the completion time from 30.55 to 5.37 seconds per frame (>5x) on SemanticKITTI and achieves superior performance compared to state-of-the-art 3D LiDAR scene completion models. Our model and code are publicly available on https://github.com/happyw1nd/ScoreLiDAR. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2412_03515 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Distilling Diffusion Models to Efficient 3D LiDAR Scene Completion Zhang, Shengyuan Zhao, An Yang, Ling Li, Zejian Meng, Chenye Xu, Haoran Chen, Tianrun Wei, AnYang GU, Perry Pengyun Sun, Lingyun Computer Vision and Pattern Recognition Diffusion models have been applied to 3D LiDAR scene completion due to their strong training stability and high completion quality. However, the slow sampling speed limits the practical application of diffusion-based scene completion models since autonomous vehicles require an efficient perception of surrounding environments. This paper proposes a novel distillation method tailored for 3D Li- DAR scene completion models, dubbed ScoreLiDAR, which achieves efficient yet high-quality scene completion. Score- LiDAR enables the distilled model to sample in significantly fewer steps after distillation. To improve completion quality, we also introduce a novel Structural Loss, which encourages the distilled model to capture the geometric structure of the 3D LiDAR scene. The loss contains a scene-wise term constraining the holistic structure and a point-wise term constraining the key landmark points and their relative configuration. Extensive experiments demonstrate that ScoreLiDAR significantly accelerates the completion time from 30.55 to 5.37 seconds per frame (>5x) on SemanticKITTI and achieves superior performance compared to state-of-the-art 3D LiDAR scene completion models. Our model and code are publicly available on https://github.com/happyw1nd/ScoreLiDAR. |
| title | Distilling Diffusion Models to Efficient 3D LiDAR Scene Completion |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2412.03515 |