Radish: Radar Implicit Shapes Via Extreme-Wideband Sub-Terahertz Synthetic Aperture Radar
We explore synthetic aperture radar (SAR) 3D imaging capabilities in the sub-THz band using a novel 110-260 GHz experimental testbed. We propose RADar Implicit SHapes (RADISH), a post-processing method that leverages the SAR’s millimeter-level imaging resolution to estimate an object’s 3D shape. RADISH first c onverts high-resolution SAR images to a detailed point cloud of a scanned object. The point cloud is then used to fit an implicit neural representation to the object’s surface by approximating the signed distance function of the scene. We experimentally validate RADISH’s ability to represent the salient geometric features of real-world 3D objects.
Authors
- Edwin Vargas (ORCID: https://orcid.org/0000-0002-7979-9497)
- Sean M. Farrell (ORCID: https://orcid.org/0009-0000-8424-1968)
- Brian Woods (ORCID: https://orcid.org/0000-0003-4363-4444)
- Ariel Habshush
- Ashutosh Sabharwal (ORCID: https://orcid.org/0000-0003-1898-5787)
- Ashok Veeraraghavan (ORCID: https://orcid.org/0000-0001-5043-7460)
- Henry O. Everitt (ORCID: https://orcid.org/0000-0002-8141-3768)
- Zhang Li (ORCID: https://orcid.org/0009-0003-8559-4599)
- Kris Li
Institutions
- United States Army Medical Command (US)
- United States Department of the Army (US)
- MaXentric Technologies (United States) (US)
- Rice University (US)
Publication Details
- Published
- 2025-06-02
- DOI
- https://doi.org/10.1109/cisa64343.2025.11091563
- Primary Topic
- Antenna Design and Optimization
- Type
- article
- Field-Weighted Citation Impact
- 0.00
Funders
- National Science Foundation
- Arm