Abstract
Point cloud scene flow estimation is of practical importance for dynamic scene navigation in autonomous driving. Since scene flow labels are hard to obtain, current methods train their models on synthetic data and transfer them to real scenes. However, large disparities between existing synthetic datasets and real scenes lead to poor model transfer. We make two major contributions to address that. First, we develop a point cloud collector and scene flow annotator for GTA-V engine to automatically obtain diverse realistic training samples without human intervention. With that, we develop a large-scale synthetic scene flow dataset GTA-SF. Second, we propose a mean-teacher-based domain adaptation framework that leverages self-generated pseudo-labels of the target domain. It also explicitly incorporates shape deformation regularization and surface correspondence refinement to address distortions and misalignments in domain transfer. Through extensive experiments, we show that our GTA-SF dataset leads to a consistent boost in model generalization to three real datasets (i.e., Waymo, Lyft and KITTI) as compared to the most widely used FT3D dataset. Moreover, our framework achieves superior adaptation performance on six source-target dataset pairs, remarkably closing the average domain gap by 60%. Data and codes are available at https://github.com/leolyj/DCA-SRSFE
Original language | English |
---|---|
Title of host publication | Proceedings - 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, CVPR 2022 |
Editors | Kosta Derpanis |
Place of Publication | Piscataway NJ USA |
Publisher | IEEE, Institute of Electrical and Electronics Engineers |
Pages | 7223-7233 |
Number of pages | 11 |
ISBN (Electronic) | 9781665469463 |
ISBN (Print) | 9781665469470 |
DOIs | |
Publication status | Published - 2022 |
Event | IEEE Conference on Computer Vision and Pattern Recognition 2022 - New Orleans, United States of America Duration: 19 Jun 2022 → 24 Jun 2022 https://ieeexplore.ieee.org/xpl/conhome/9878378/proceeding (Proceedings) https://cvpr2022.thecvf.com https://cvpr2022.thecvf.com/ (Website) |
Conference
Conference | IEEE Conference on Computer Vision and Pattern Recognition 2022 |
---|---|
Abbreviated title | CVPR 2022 |
Country/Territory | United States of America |
City | New Orleans |
Period | 19/06/22 → 24/06/22 |
Internet address |
Keywords
- Deep learning architectures and techniques
- Motion and tracking
- Robot vision
- Scene analysis and understanding
- Transfer/low-shot/long-tail learning