Enhancing Transformer-Based RGB-T Image Fusion via Hybrid Pixel Upsampling
4th IEEE Conference on Information Technology and Data Science, CITDS 2026, Hybrid, Debrecen, Macaristan, 27 - 28 Ağustos 2026, ss.319-324, (Tam Metin Bildiri)
- Yayın Türü: Bildiri / Tam Metin Bildiri
- Doi Numarası: 10.1109/citds70571.2026.11706909
- Basıldığı Şehir: Hybrid, Debrecen
- Basıldığı Ülke: Macaristan
- Sayfa Sayıları: ss.319-324
- Anahtar Kelimeler: convolutional neural networks, deep learning, guided thermal super-resolution, image fusion, thermal image reconstruction, upsampling
- Ankara Üniversitesi Adresli: Evet
Özet
Guided thermal super-resolution (GTSR) is a cross-modal deep learning task that reconstructs high-resolution thermal images from low-resolution inputs using structural guidance from high-resolution RGB images. This problem is relevant to intelligent systems and data-driven applications where highresolution thermal sensors are limited by hardware constraints. Recent transformer-based fusion models achieve strong cross-modal performance but rely on fixed bicubic interpolation during early upsampling, which can degrade high-frequency spatial details prior to feature extraction. In this work, a hybrid upsampling module is proposed that combines learnable PixelShufflebased feature expansion with a bicubic residual skip connection, integrated into a SwinFuSR-based pipeline without modifying the backbone architecture. Experiments on the CIDIS dataset at scale factor × 2 show that the proposed method achieves comparable PSNR to Lanczos interpolation on the validation set (44.302 vs 44.305 dB), with a statistically significant but marginal advantage on the held-out test set (p=0.029, paired t-test, n=100), while maintaining low computational overhead (+1.5% GFLOPs). An ablation study confirms that the bicubic residual skip is the dominant contributor, with the learnable PixelShuffle module providing a consistent additional improvement.