ShearFuse-UNet: Hadamard, DCT, and Shearlet Transform Fusion for Next-Day Wildfire Spread Prediction
Quick Answer
This paper shows that The ShearFuse-UNet model integrates Hadamard, DCT, and Shearlet transforms for efficient next-day wildfire spread prediction, achieving an F1 score of 0.596 with only 267k parameters, outperforming a ResNet18-based U-Net.
Quick Take
The ShearFuse-UNet model integrates Hadamard, DCT, and Shearlet transforms for efficient next-day wildfire spread prediction, achieving an F1 score of 0.596 with only 267k parameters, outperforming a ResNet18-based U-Net. This model demonstrates a favorable accuracy-efficiency trade-off, validated on multiple datasets.
Key Points
- ShearFuse-UNet uses a 2D Fast Walsh-Hadamard Transform, DCT, and Shearlet branches.
- Achieves an F1 score of 0.596 with only 267k parameters.
- Outperforms ResNet18-based U-Net, which has 14M parameters and F1 = 0.589.
- Demonstrates efficient accuracy trade-off validated on the WildfireSpreadTS dataset.
- The model relies on fixed transforms, reducing computational costs.
Paper Resources
Article Content
From source RSS / original summaryarXiv:2606. 14071v1 Announce Type: new Abstract: We propose ShearFuse-UNet, a lightweight and computationally efficient deep learning model for next-day wildfire spread prediction from multi-modal satellite data. The model integrates three complementary transform-domain branches inside each encoder block of a U-Net backbone: a 2D Fast Walsh-Hadamard Transform (WHT) branch, a 2D Discrete Cosine Transform (DCT) branch, and a cone-adapted digital Shearlet residual branch.
The WHT and DCT branches establish orthogonal latent spaces with learnable spectral scaling and fixed soft-thresholding, while the Shearlet branch provides anisotropic, multi-directional feature decomposition that explicitly encodes the elongated edge structures characteristic of fire fronts. A learned SpectralFusion gate adaptively combines the WHT and DCT responses, and the Shearlet reconstruction is added as a residual.
This three-branch design bears a loose structural analogy to transformer self-attention: the WHT and DCT branches provide complementary spectral representations that are adaptively fused, while the Shearlet branch contributes directional content through a residual pathway. Unlike self-attention, the proposed design relies on fixed mathematical transforms rather than learned projection operators, reducing parameter count and computational cost.
Evaluated on the WildfireSpreadTS dataset, ShearFuse-UNet achieves an F1 score of 0. 596 with only 267k parameters, outperforming a ResNet18-based U-Net (14M parameters, F1 = 0. 589) and demonstrating a highly favorable accuracy-efficiency trade-off. Results on the Google Next-Day Wildfire Spread dataset further validate these findings across a different benchmark.
Reader Mode unavailable (could not extract clean content).
Want this in your inbox every morning?
Daily brief at your local 8am — bilingual EN/中文, free.
More from arXiv cs.CV
See more →LLM-Guided ANN Index Optimization for Human-Object Interaction Retrieval
A phase-aware LLM agent optimizes human-object interaction retrieval, outperforming Optuna TPE by 33.3% and VDTuner by 34.2% on the HICO-DET benchmark. This method enhances throughput by 15.3x over UniIR and demonstrates strong transferability across vector database management systems.