HVDistill: Transferring Knowledge from Images to Point Clouds via Unsupervised Hybrid-View Distillation
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866914719489589248 |
|---|---|
| author | Zhang, Sha Deng, Jiajun Bai, Lei Li, Houqiang Ouyang, Wanli Zhang, Yanyong |
| author_facet | Zhang, Sha Deng, Jiajun Bai, Lei Li, Houqiang Ouyang, Wanli Zhang, Yanyong |
| contents | We present a hybrid-view-based knowledge distillation framework, termed HVDistill, to guide the feature learning of a point cloud neural network with a pre-trained image network in an unsupervised manner. By exploiting the geometric relationship between RGB cameras and LiDAR sensors, the correspondence between the two modalities based on both image-plane view and bird-eye view can be established, which facilitates representation learning. Specifically, the image-plane correspondences can be simply obtained by projecting the point clouds, while the bird-eye-view correspondences can be achieved by lifting pixels to the 3D space with the predicted depths under the supervision of projected point clouds. The image teacher networks provide rich semantics from the image-plane view and meanwhile acquire geometric information from the bird-eye view. Indeed, image features from the two views naturally complement each other and together can ameliorate the learned feature representation of the point cloud student networks. Moreover, with a self-supervised pre-trained 2D network, HVDistill requires neither 2D nor 3D annotations. We pre-train our model on nuScenes dataset and transfer it to several downstream tasks on nuScenes, SemanticKITTI, and KITTI datasets for evaluation. Extensive experimental results show that our method achieves consistent improvements over the baseline trained from scratch and significantly outperforms the existing schemes. Codes are available at git@github.com:zhangsha1024/HVDistill.git. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2403_11817 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | HVDistill: Transferring Knowledge from Images to Point Clouds via Unsupervised Hybrid-View Distillation Zhang, Sha Deng, Jiajun Bai, Lei Li, Houqiang Ouyang, Wanli Zhang, Yanyong Computer Vision and Pattern Recognition We present a hybrid-view-based knowledge distillation framework, termed HVDistill, to guide the feature learning of a point cloud neural network with a pre-trained image network in an unsupervised manner. By exploiting the geometric relationship between RGB cameras and LiDAR sensors, the correspondence between the two modalities based on both image-plane view and bird-eye view can be established, which facilitates representation learning. Specifically, the image-plane correspondences can be simply obtained by projecting the point clouds, while the bird-eye-view correspondences can be achieved by lifting pixels to the 3D space with the predicted depths under the supervision of projected point clouds. The image teacher networks provide rich semantics from the image-plane view and meanwhile acquire geometric information from the bird-eye view. Indeed, image features from the two views naturally complement each other and together can ameliorate the learned feature representation of the point cloud student networks. Moreover, with a self-supervised pre-trained 2D network, HVDistill requires neither 2D nor 3D annotations. We pre-train our model on nuScenes dataset and transfer it to several downstream tasks on nuScenes, SemanticKITTI, and KITTI datasets for evaluation. Extensive experimental results show that our method achieves consistent improvements over the baseline trained from scratch and significantly outperforms the existing schemes. Codes are available at git@github.com:zhangsha1024/HVDistill.git. |
| title | HVDistill: Transferring Knowledge from Images to Point Clouds via Unsupervised Hybrid-View Distillation |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2403.11817 |