PINN-Kerr
A physics-informed neural network that solves the vacuum Einstein equations for the Kerr metric, coupled with a geodesic ray tracer that renders the resulting black hole.
Research and space-industry work.
A physics-informed neural network that solves the vacuum Einstein equations for the Kerr metric, coupled with a geodesic ray tracer that renders the resulting black hole.
A learnable lifting transform that splits manifold-valued signals into coarse and detail components, with an explicit inverse.
Classical methods vs. deep learning for satellite change detection, benchmarked on four public datasets.
Multi-task building and vegetation segmentation plus height maps from geospatial foundation-model embeddings.
Signal filtering and anomaly detection on data from the GNOME network of optical magnetometers, searching for candidate transient events in dark-matter searches.
Detection, segmentation and GAN-based inpainting of foreign-body artefacts in chest X-rays, and their impact on aortic calcification scoring.