Imaging Interiors: An Implicit Solution to Electromagnetic Inverse Scattering Problems

This is the repository that contains source code for the project page of Imaging Interiors: An Implicit Solution to Electromagnetic Inverse Scattering Problems.

Quick Start

To run a case of the Austria profile with a size of 64x64, just follow the following instructions.

Data preparation

The data is generated based on this GitHub repository. You can download the data for the Austria profile from this Google Drive link and place the data under the ./data/ folder.

Run the code

Execute the following command:

python run_isp_double_MLP_L1.py --config configs/circle_double_MLP_Au_L1.txt

If you find our paper useful for your work please cite:

@inproceedings{luo2024imaging,
  author    = {Ziyuan Luo and Boxin Shi and Haoliang Li and Renjie Wan},
  title     = {Imaging Interiors: An Implicit Solution to Electromagnetic Inverse Scattering Problems},
  booktitle   = {European Conference on Computer Vision},
  year      = {2024},
}