Finetuning Pre-trained Model with Limited Data for LiDAR-based 3D Object Detection by Bridging Domain Gaps
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2024
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866912055383031808 |
|---|---|
| author | Jang, Jiyun Chang, Mincheol Park, Jongwon Kim, Jinkyu |
| author_facet | Jang, Jiyun Chang, Mincheol Park, Jongwon Kim, Jinkyu |
| contents | LiDAR-based 3D object detectors have been largely utilized in various applications, including autonomous vehicles or mobile robots. However, LiDAR-based detectors often fail to adapt well to target domains with different sensor configurations (e.g., types of sensors, spatial resolution, or FOVs) and location shifts. Collecting and annotating datasets in a new setup is commonly required to reduce such gaps, but it is often expensive and time-consuming. Recent studies suggest that pre-trained backbones can be learned in a self-supervised manner with large-scale unlabeled LiDAR frames. However, despite their expressive representations, they remain challenging to generalize well without substantial amounts of data from the target domain. Thus, we propose a novel method, called Domain Adaptive Distill-Tuning (DADT), to adapt a pre-trained model with limited target data (approximately 100 LiDAR frames), retaining its representation power and preventing it from overfitting. Specifically, we use regularizers to align object-level and context-level representations between the pre-trained and finetuned models in a teacher-student architecture. Our experiments with driving benchmarks, i.e., Waymo Open dataset and KITTI, confirm that our method effectively finetunes a pre-trained model, achieving significant gains in accuracy. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_01319 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Finetuning Pre-trained Model with Limited Data for LiDAR-based 3D Object Detection by Bridging Domain Gaps Jang, Jiyun Chang, Mincheol Park, Jongwon Kim, Jinkyu Computer Vision and Pattern Recognition Artificial Intelligence Robotics LiDAR-based 3D object detectors have been largely utilized in various applications, including autonomous vehicles or mobile robots. However, LiDAR-based detectors often fail to adapt well to target domains with different sensor configurations (e.g., types of sensors, spatial resolution, or FOVs) and location shifts. Collecting and annotating datasets in a new setup is commonly required to reduce such gaps, but it is often expensive and time-consuming. Recent studies suggest that pre-trained backbones can be learned in a self-supervised manner with large-scale unlabeled LiDAR frames. However, despite their expressive representations, they remain challenging to generalize well without substantial amounts of data from the target domain. Thus, we propose a novel method, called Domain Adaptive Distill-Tuning (DADT), to adapt a pre-trained model with limited target data (approximately 100 LiDAR frames), retaining its representation power and preventing it from overfitting. Specifically, we use regularizers to align object-level and context-level representations between the pre-trained and finetuned models in a teacher-student architecture. Our experiments with driving benchmarks, i.e., Waymo Open dataset and KITTI, confirm that our method effectively finetunes a pre-trained model, achieving significant gains in accuracy. |
| title | Finetuning Pre-trained Model with Limited Data for LiDAR-based 3D Object Detection by Bridging Domain Gaps |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence Robotics |
| url | https://arxiv.org/abs/2410.01319 |