Dual-Constrained Diffusion Image Compression for Operational Rate-Distortion-Perception Optimization
Researchers have developed a novel image compression technique called Dual-Constrained Diffusion Image Compression (DCIC). This method integrates a learned codec with a diffusion-based decoder, utilizing distortion and idempotence constraints to jointly optimize for fidelity and perceptual realism. DCIC allows for continuous adjustment of the trade-off between these two aspects from a single bitstream, offering superior performance in both fidelity and perceptual quality compared to existing methods. AI
IMPACT This research could lead to more efficient and perceptually pleasing image compression techniques, impacting media storage and transmission.