ML Anthology

Loop Mining Large-Scale Unlabeled Data for Corner Case Detection in Autonomous Driving

Jiawei Zhao, Yiting Duan, Jinming Su, Wangwang Yang, Tingyi Guo, Xingyue Chen, Junfeng Luo
ECCVW 2024 pp. 277-286
doi:10.1007/978-3-031-91767-7_19 /eccvw/2024/zhao2024eccvw-loop/

Abstract

For obstacle detection in road scenes, it is very challenging to detect novel objects that are not seen or barely seen during training. To address this issue, we propose an efficient pipeline for obstacle detection in road scenes based on large-scale unlabeled data. Specifically, we use large-scale unlabeled data to train a closed-set model and a open-set model separately in a pseudo-supervised learning manner, and then iteratively improve the performance of both models through the proposed loop-optimization strategy, which employs some useful tricks to remove false positive detections about corner cases. Experimental evidence demonstrates that our approach achieves new state-of-the-art on the popular CODA dataset.

PDF ECCVW Semantic Scholar

Cite

Text

Zhao et al. "Loop Mining Large-Scale Unlabeled Data for Corner Case Detection in Autonomous Driving." European Conference on Computer Vision Workshops, 2024. doi:10.1007/978-3-031-91767-7_19

Markdown

[Zhao et al. "Loop Mining Large-Scale Unlabeled Data for Corner Case Detection in Autonomous Driving." European Conference on Computer Vision Workshops, 2024.](https://mlanthology.org/eccvw/2024/zhao2024eccvw-loop/) doi:10.1007/978-3-031-91767-7_19

BibTeX

@inproceedings{zhao2024eccvw-loop,
  title     = {{Loop Mining Large-Scale Unlabeled Data for Corner Case Detection in Autonomous Driving}},
  author    = {Zhao, Jiawei and Duan, Yiting and Su, Jinming and Yang, Wangwang and Guo, Tingyi and Chen, Xingyue and Luo, Junfeng},
  booktitle = {European Conference on Computer Vision Workshops},
  year      = {2024},
  pages     = {277-286},
  doi       = {10.1007/978-3-031-91767-7_19},
  url       = {https://mlanthology.org/eccvw/2024/zhao2024eccvw-loop/}
}