A single framework that first fuses bracketed exposures into a clean SDR via diffusion with dual cross-attention, then predicts a gain map and global scale to lift that SDR into an HDR image.
In: AAAI (2023)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Dual-Output Multi-Exposure HDR Reconstruction via SDR Fusion and Gain Map Inverse Tone Mapping
A single framework that first fuses bracketed exposures into a clean SDR via diffusion with dual cross-attention, then predicts a gain map and global scale to lift that SDR into an HDR image.