3 papers
cs.LG2025
Semantic-Aware Gaussian Process Calibration with Structured Layerwise Kernels for Deep Neural Networks
Kyung-hwan Lee, Kyung-tae Kim
Calibrating the confidence of neural network classifiers is essential for quantifying the reliability of their predictions during inference. However, conventional Gaussian Process…
cs.CV2024
IRASNet: Improved Feature-Level Clutter Reduction for Domain Generalized SAR-ATR
Oh-Tae Jang, Min-Jun Kim, Sung-Ho Kim +2
Recently, computer-aided design models and electromagnetic simulations have been used to augment synthetic aperture radar (SAR) data for deep learning. However, an automatic target…
cs.CV2024
Adaptive Residual Transformation for Enhanced Feature-Based OOD Detection in SAR Imagery
Kyung-hwan Lee, Kyung-tae Kim
Recent advances in deep learning architectures have enabled efficient and accurate classification of pre-trained targets in Synthetic Aperture Radar (SAR) images. Nevertheless, the…