Back to feed
arXiv cs.LG
arXiv cs.LG
7/17/2026
QFireNet: A Quantum-Enhanced U-Net for Wildfire Segmentation from Sentinel-2 Imagery

QFireNet: A Quantum-Enhanced U-Net for Wildfire Segmentation from Sentinel-2 Imagery

Short summary

QFireNet integrates a variational quantum circuit into the U-Net bottleneck for wildfire segmentation from Sentinel-2 imagery. Quantum-hybrid variants QB-Net (F1=31.18) and QuFeX (F1=30.79) outperformed the classical U-Net baseline (F1=28.71), though a classical FPN with data mixing reached F1=39.76 by removing domain shift. Results suggest quantum ML may offer advantages for high-dimensional spectral feature spaces in remote sensing.

  • •Quantum-hybrid U-Net variants outperform classical baseline on wildfire segmentation
  • •Data mixing to remove domain shift boosted classical FPN to F1=39.76, surpassing quantum models
  • •Cross-dataset transfer on CaBuAr validates architecture robustness and generalizability

Generated with AI, which can make mistakes.

#research-breakthrough#open-source
Read full article at arXiv cs.LG

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

QANTIS: Hardware-Calibrated Sequential POMDP Belief Updates on IBM Heron
arXiv CS.AI
QuantFlow: A Federated Mamba-Based Post-Transformer Foundation Model for Time-Series Forecasting
arXiv cs.LG
The original title is "I built AutoML for quantum machine learning — here's the architecture"
Dev.to
From hyperplanes to hyperellipsoids: characterizing the inherent interpretability of linear and single-qubit mixed-state binary classification models
arXiv cs.LG