Riestiya Zain Fadillah
Publikasjoner
Uncertainty Quantification in Video Distortion Classification Under Dataset Shift
Assessing SoTA video distortion classification model performance and uncertainty (using Monte Carlo Dropout and uncertainty metrics) under concept and covariate shift.
Automated thread counting in archaeological textiles using pulse-based feature extraction
A framework for automating thread counting in textile images using a feature extractor to capture thread counts in vertical and horizontal directions as a pulse-counting problem, inspired by how humans intuitively count threads.
Few-shot Supervised Contrastive Learning for Image/Video Distortion Classification
A new Supervised Contrastive (SupCon) framework is proposed for image/video distortion classification through the novel creation of contrastive views for the encoder to effectively learn the distortion manifold using only a limited amount of labeled data.