Enhancing WISE Infrared Imaging to Spitzer Resolution Using Deep Learning Super-Resolution
Abstract
We present a deep-learning framework that performs 4.6x spatial super-resolution from WISE W1 (3.4 micron) toward Spitzer IRAC Ch1 (3.6 micron), and characterize its behavior on the COSMOS field. We train on ~390,000 paired cutouts drawn uniformly within the WISE/Spitzer overlap and report all metrics on a held-out test set of 83,592 cutouts. The framework uses a convolutional neural network trained with a loss function that emphasizes accurate recovery of sources in crowded fields. The model recovers the total flux of the central source in a fixed aperture to a median relative error of 11%, a factor of ~2 better than the interpolation baselines; the gain reaches ~3x on the faintest quartile. The brightness dependence is monotonic: the aperture integrated flux error decreases from 13% on the faintest quartile to 8% on the brightest. At the 3-5 arcsec separations where WISE blends sources that Spitzer separates, the model recovers 35% of the source peaks detectable in the Spitzer truth compared with 9% for interpolation. The characteristic failure mode is oversmoothing of source profiles, which biases integrated flux measurements upward; this pattern is qualitatively similar to that of the interpolation baselines but is quantitatively smaller for the trained model. These results suggest genuine resolution enhancement and source deblending, providing a path toward applying super-resolution across the all-sky area that Spitzer could not cover. An appendix replicates the analysis at W2 -> IRAC Ch2 with consistent results; the trained model and code are publicly available.
Related
- The Spitzer Data Fusion - A Far-Ultraviolet to Far-Infrared Multi-Wavelength Database in Spitzer Extragalactic Survey Fields
- Enhancing the Angular Resolution of Large Array of imaging atmospheric Cherenkov Telescope (LACT) at Ultra-High Energies
- Tracing star formation across the Rosette nebula using deep Spitzer imaging
- Ushering in sub-arcsecond angular resolution observatories in the far-infrared: The case for a double-Fourier interferometer
- LIGHTS: Limitations of far-infrared tracers for Galactic cirrus mitigation in the era of ultra-deep imaging
BibTeX
@misc{rezaee2026enhancing,
title = {{Enhancing WISE Infrared Imaging to Spitzer Resolution Using Deep Learning Super-Resolution}},
author = {Saeed Rezaee and Shoubaneh Hemmati and Bahram Mobasher and Naveen A. Reddy and Tara Fetherolf and Alexander de la Vega and Negin Moharrami Allafi and Sina Miri},
howpublished = {arXiv},
year = {2026},
month = sep,
doi = {10.48550/arxiv.2607.14295},
eprint = {2607.14295},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2607.14295},
}
RIS
TY - UNPB
TI - Enhancing WISE Infrared Imaging to Spitzer Resolution Using Deep Learning Super-Resolution
AU - Saeed Rezaee
AU - Shoubaneh Hemmati
AU - Bahram Mobasher
AU - Naveen A. Reddy
AU - Tara Fetherolf
AU - Alexander de la Vega
AU - Negin Moharrami Allafi
AU - Sina Miri
PY - 2026
DA - 2026/09/09
PB - arXiv
DO - 10.48550/arxiv.2607.14295
UR - https://arxiv.org/abs/2607.14295
AB - We present a deep-learning framework that performs 4.6x spatial super-resolution from WISE W1 (3.4 micron) toward Spitzer IRAC Ch1 (3.6 micron), and characterize its behavior on the COSMOS field. We train on ~390,000 paired cutouts drawn uniformly within the WISE/Spitzer overlap and report all metrics on a held-out test set of 83,592 cutouts. The framework uses a convolutional neural network trained with a loss function that emphasizes accurate recovery of sources in crowded fields. The model recovers the total flux of the central source in a fixed aperture to a median relative error of 11%, a factor of ~2 better than the interpolation baselines; the gain reaches ~3x on the faintest quartile. The brightness dependence is monotonic: the aperture integrated flux error decreases from 13% on the faintest quartile to 8% on the brightest. At the 3-5 arcsec separations where WISE blends sources that Spitzer separates, the model recovers 35% of the source peaks detectable in the Spitzer truth compared with 9% for interpolation. The characteristic failure mode is oversmoothing of source profiles, which biases integrated flux measurements upward; this pattern is qualitatively similar to that of the interpolation baselines but is quantitatively smaller for the trained model. These results suggest genuine resolution enhancement and source deblending, providing a path toward applying super-resolution across the all-sky area that Spitzer could not cover. An appendix replicates the analysis at W2 -> IRAC Ch2 with consistent results; the trained model and code are publicly available.
ER -