Code

Topology-aware optimal transport for source-free test-time anomaly segmentation.

Persistent-homology pseudo-labels for test-time adaptation in anomaly segmentation.

Hypergraph contrastive fusion of vibration signals and spectrograms for motor fault diagnosis.

RGB and point cloud fusion with attention-guided restoration for industrial anomaly detection.

Relative depth from a single hyperspectral image using chromatic aberration and defocus blur.

Early detection of sugarcane smut disease from hyperspectral images.

Detecting partly hidden objects using spectral cues, from the PhD thesis.

Material-based boundary detection in hyperspectral images.

3D reconstruction that combines the geometry recovered from each spectral band.

Relative depth from band-to-band changes in defocus blur.

Shared functions for reading, correcting, normalising and viewing hyperspectral cubes.

Python packages for machine learning on topological domains: TopoNetX, TopoEmbedX and TopoModelX. I am one of the contributors.

Site and interactive filtration demo for our topological deep learning tutorial. site