ListarInstitute of New Imaging Technologies (INIT) por tema "task analysis"
Mostrando ítems 1-3 de 3
-
Noise-Tolerant Deep Neighborhood Embedding for Remotely Sensed Images With Label Noise
IEEE (2021-02-02)Recently, many deep learning-based methods have been developed for solving remote sensing (RS) scene classification or retrieval tasks. Most of the adopted loss functions for training these models require accurate annotations. ... -
Revisiting the Dissimilarity Representation in the Context of Regression
IEEE (2021-12-02)In machine learning, a natural way to represent an instance is by using a feature vector. However, several studies have shown that this representation may not accurately characterize an object. For classification problems, ...
-
Rotation-Invariant Deep Embedding for Remote Sensing Images
Institute of Electrical and Electronics Engineers (2021-06-28)Endowing convolutional neural networks (CNNs) with the rotation-invariant capability is important for characterizing the semantic contents of remote sensing (RS) images since they do not have typical orientations. Most of ...
