MagicFuse derives a cross-spectral scene representation from one degraded visible image via intra-spectral reinforcement and cross-spectral generation diffusion branches whose probabilistic outputs are fused under visual and semantic constraints.
Multi-interactive feature learning and a full-time multi-modality benchmark for image fusion and segmentation
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
MagicFuse: Single Image Fusion for Visual and Semantic Reinforcement
MagicFuse derives a cross-spectral scene representation from one degraded visible image via intra-spectral reinforcement and cross-spectral generation diffusion branches whose probabilistic outputs are fused under visual and semantic constraints.