Contrastive Learning for Natural Language-Based Vehicle Retrieval
Abstract
AI City Challenge 2021 Task 5: The Natural Language-Based Vehicle Tracking is a Natural Language-based Vehicle Retrieval task, which requires retrieving a single-camera track using a set of three natural language descriptions of the specific targets. In this paper, we present our methods to tackle the difficulties of the provided task. Experiments with our approaches on the competitive dataset from AICity Challenge 2021 show that our techniques achieve Mean Reciprocal Rank score of 0.1701 on the public test dataset and 0.1571 on the private test dataset.
Cite
Text
Nguyen et al. "Contrastive Learning for Natural Language-Based Vehicle Retrieval." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021. doi:10.1109/CVPRW53098.2021.00480Markdown
[Nguyen et al. "Contrastive Learning for Natural Language-Based Vehicle Retrieval." IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, 2021.](https://mlanthology.org/cvprw/2021/nguyen2021cvprw-contrastive/) doi:10.1109/CVPRW53098.2021.00480BibTeX
@inproceedings{nguyen2021cvprw-contrastive,
title = {{Contrastive Learning for Natural Language-Based Vehicle Retrieval}},
author = {Nguyen, Tam Minh and Huu, Quang Pham and Bao, Linh Doan and Trinh, Hoang Viet and Nguyen, Viet Anh and Phan, Viet-Hoang},
booktitle = {IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops},
year = {2021},
pages = {4245-4252},
doi = {10.1109/CVPRW53098.2021.00480},
url = {https://mlanthology.org/cvprw/2021/nguyen2021cvprw-contrastive/}
}