Spatial Signal Processing tries to draw conclusions about sources of physical fields and possibly about the geometry and physical properties of a spatial domain using underlying mathematical models and distributed observations of the field itself. The kind of problems we worked on in the past include ultrasound tomography and characterization of physical fields driven by the wave equation and the diffusion equation, and you are welcome to browse pages of past project to find more details.
Currently, we put more focus on localizing acoustic and light sources, inferring room geometry, or even going a step further to combine acoustic localization and mapping into a single procedure — the infamous acoustic SLAM.
Recent LCAV publications in this area:
DeepWave: A Recurrent Neural-Network for Real-Time Acoustic Imaging2019-12-16. Thirty-third Conference on Neural Information Processing Systems (NeurIPS), Vancouver, British Columbia, Canada, December 9-14, 2019.
Towards Real-Time High-Resolution Interferometric Imaging with Bluebild2017-08-01.
Look, no beacons! Optimal all-in-one EchoSLAM2016. 50th Asilomar Conference on Signals, Systems, and Computers, Asilomar, Pacific Grove, CA, November 6-9, 2016.
How to Localize Ten Microphones in One Fingersnap2014. 22nd European Signal Processing Conference, Lisbon, Portugal, September 1-5, 2014. p. 2275-2279.