Atanasova et al. (2026) A Multitask Framework with Cross-task Selective Feature Sharing for Remote Sensing Image Time Series Analysis

Identification

• Journal: IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
• Year: 2026
• Date: 2026-01-01
• Authors: Milena Atanasova, Luca Bergamasco, Francesca Bovolo
• DOI: 10.1109/jstars.2026.3673598

Research Groups

[Not provided in the input text]

Short Summary

This paper introduces a multitask framework that employs cross-task selective feature sharing to enhance the analysis of remote sensing image time series.

Objective

• To develop and evaluate a novel multitask framework incorporating cross-task selective feature sharing for improved analysis of remote sensing image time series.

Study Configuration

• Spatial Scale: Implied to be regional to global, depending on the remote sensing data used, but not specified in the input text.
• Temporal Scale: Implied to be multi-temporal, covering a time series, but not specified in the input text.

Methodology and Data

• Models used: A multitask learning framework with cross-task selective feature sharing. Specific model architectures (e.g., deep learning models) are not detailed in the provided text.
• Data sources: Remote sensing image time series. Specific satellite missions, sensors, or datasets are not detailed in the provided text.

Main Results

[Not provided in the input text]

Contributions

[Not provided in the input text]

Funding

[Not provided in the input text]

Citation

@article{Atanasova2026Multitask,
  author = {Atanasova, Milena and Bergamasco, Luca and Bovolo, Francesca},
  title = {A Multitask Framework with Cross-task Selective Feature Sharing for Remote Sensing Image Time Series Analysis},
  journal = {IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing},
  year = {2026},
  doi = {10.1109/jstars.2026.3673598},
  url = {https://doi.org/10.1109/jstars.2026.3673598}
}