Back to datasets
Dataset assetOpen Source CommunityImage DatasetUrban Flood Detection

UW-Bench

The UW‑Bench dataset, created by the School of Microelectronics and Communication Engineering at Chongqing University, focuses on urban flood detection, containing 7,677 annotated images sourced from surveillance cameras and handheld devices. The dataset was collected under various adverse conditions such as low light, strong reflections, and clear water, aiming to improve model generalization in real‑world applications. Manual annotation ensures data quality, suitable for enhancing accuracy and efficiency of urban flood detection.

Source
arXiv
Created
Jul 11, 2024
Updated
Jul 11, 2024
Signals
679 views
Availability
Linked source ready
Overview

Dataset description and usage context

Urban Flood Detection Dataset (UW‑Bench)

Overview

  • Dataset Name: Urban Waterlogging Benchmark (UW‑Bench)
  • Purpose: For urban flood detection, especially under diverse and adverse environmental conditions.
  • Features: Includes regular samples and challenging samples to objectively assess model capability in real‑world scenarios.

Details

  • Training Set: Collected and annotated by the LiVE team at Chongqing University.
  • Test Set: Provided by Huawei.
  • Dataset Links:

Citation

@inproceedings{ song2024lsmadapter, title={Urban Waterlogging Detection: A Challenging Benchmark and Large‑Small Model Co‑Adapter}, author={Suqi Song and Chenxu Zhang and Peng Zhang and Pengkun Li and Fenglong Song and Lei Zhang}, journal = {ECCV}, issue_date = {2024} }

Need downstream help?

Pair the dataset with AI analysis and content workflows.

Once the source passes your review, move straight into summarization, transformation, report drafting, or presentation generation with the JuheAI toolchain.

Explore AI studio