EvaNet: Elevation-Guided Flood Extent Mapping on Earth Imagery

Authors: Mirza Tanzim Sami, Da Yan, Saugat Adhikari, Lyuheng Yuan, Jiao Han, Zhe Jiang, Jalal Khalil, Yang Zhou

IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments show that Eva Net significantly outperforms the U-Net baselines, and works as a perfect drop-in replacement for U-Net in existing solutions to flood extent mapping.
Researcher Affiliation Academia 1Indiana University Bloomington 2University of Alabama at Birmingham 3University of Florida 4St. Cloud State University 5Auburn University
Pseudocode No The paper describes the formulations and architecture (e.g., Figure 4), but it does not contain any explicitly labeled pseudocode or algorithm blocks.
Open Source Code Yes Eva Net is open-sourced at https://github.com/MTSami/Eva Net.
Open Datasets Yes We obtain high-resolution aerial imagery from NOAA National Geodetic Survey during Hurricane Matthew in North Carolina (NC) in 2016 [NOAA, 2016]. The accompanied DEM data are obtained from the University of North Carolina Libraries [NCSU, 2023]... from [NOAA, 2017], and the corresponding DEM data was obtained from USGS data downloader [USGS, 2023].
Dataset Splits No Without loss of generalization, in Table 2, we use R1 and R2 for training, and use R3, R4, R5, R6 and R7 for test. The paper does not explicitly mention a separate validation split within its data partitioning description.
Hardware Specification Yes We trained our Eva Net models and the U-Net baselines on a cluster with NVIDIA-P100 GPUs for 100 epochs with a learning rate of 1e-7 and batch size of 4.
Software Dependencies No The paper mentions 'Py Torch' but does not specify version numbers for PyTorch or any other software dependencies used in the experiments.
Experiment Setup Yes We trained our Eva Net models and the U-Net baselines on a cluster with NVIDIA-P100 GPUs for 100 epochs with a learning rate of 1e-7 and batch size of 4.