Metraipolis › archive › dc_6f976307bed5
CFGPNet: Cross-Attention-Based Fused Gradient Programmed Network Framework for Multispectral Object Detection — arXiv:2608.06205
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.06205
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T10:37:45.510Z
- SHA-256 of the archived text
9605a1f53e17f3c62f2bf60f44bffcd010fdd2990f0c04b8fe50f44d0e37e657- Stable URL of this page
https://metraipolis.com/documento/dc_6f976307bed5
How to cite
CFGPNet: Cross-Attention-Based Fused Gradient Programmed Network Framework for Multispectral Object Detection — arXiv:2608.06205 — arXiv (en), https://arxiv.org/abs/2608.06205 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_6f976307bed5 (Metraipolis archive dc_6f976307bed5, sha256 9605a1f53e17f3c62f2bf60f44bffcd010fdd2990f0c04b8fe50f44d0e37e657)
BibTeX
@misc{dc6f976307bed5,
title = {CFGPNet: Cross-Attention-Based Fused Gradient Programmed Network Framework for Multispectral Object Detection — arXiv:2608.06205},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.06205},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_6f976307bed5; sha256 9605a1f53e17f3c62f2bf60f44bffcd010fdd2990f0c04b8fe50f44d0e37e657},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.06205. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Multispectral object detection combines visible and thermal imagery to improve perception under challenging illumination and environmental conditions. However, differences in modality appearance and reliability can introduce redundant or conflicting responses, limiting the use of complementary information. Complex fusion mechanisms further increase computational cost, creating a persistent trade-off between detection accuracy and efficiency. To address these challenges, CFGPNet is proposed, a cross-attention-based fused gradient programmed network. The framework incorporates re-parameterized RepViT blocks into the YOLOv9 architecture to strengthen spatial and channel representations while maintaining efficient feature extraction. Cross Computation Efficient Attention (CrossCEA) exchanges spatial attention maps between modalities at multiple detection scales, allowing each stream to emphasize regions supported by the other while preserving modality-specific information. Attention Selection and Aggregation Fusion (ASAF) combines dense feature aggregation with selection of the strongest responses from multiple attention branches to form compact, discriminative fused representations. A programmable gradient information pathway provides auxiliary supervision during training to improve feature learning. This pathway is removed after training, adding no parameters or operations at inference. Experiments on FLIR, M3FD, LLVIP, VEDAI, and MFAD demonstrate favorable accuracy-efficiency trade-offs across three model scales, with the smallest variant requiring 15.3 million parameters and 56.9 GFLOPs. The code is available at this https URL . Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-07 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.06205 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-31T10:23:31.060Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).