2025/12/06 by Wang, Kaile, He, Lijun, Fu, Haisheng +2
Computer Science · #Advanced Data Compression Techniques #Compression (physics) #Computer Vision and Pattern Recognition (cs.CV) #Data compression #FOS: Computer and information sciences #Fidelity #Focus (optics) #Generative Adversarial Networks and Image Synthesis #Generative grammar #Generative model #Image (mathematics) #Image and Video Quality Assessment #Image compression #Semantics (computer science)
paper · open access · doi:10.48550/arxiv.2512.06344
published in arXiv (Cornell University) (Cornell University)
openalex publication_date 2025/12/06 · openalex created_date 2025/12/10 · openalex updated_date 2026/07/28
Generative image compression has recently shown impressive perceptual quality, but often suffers from semantic inconsistency at ultra-low bitrates (bpp < 0.05), limiting its reliable deployment in bandwidth-constrained scenarios such as 6G semantic communications. This inconsistency stems from incomplete guidance information, which introduces semantic ambiguity into the generation process and may lead to natural-looking but source-inconsistent content. In this work, we propose a Semantic-Disambiguation-Guided Generative Image Compression (SDGIC) framework to constrain diffusion-based reconstruction at ultra-low bitrates. Specifically, SDGIC compresses the source image into three compact and complementary guidance streams: a concise text caption for global semantics, a highly compressed image (HCI) for dense visual evidence, and Reconstruction-Aware Semantic Residual Tokens (RSRTs) for reconstruction-relevant residual semantics that remain ambiguous under the text caption and HCI conditions. The RSRTs are directly optimized toward the downstream denoising objective, enabling them to provide source-specific semantic constraints for disambiguating diffusion-based reconstruction. To inject these three guidance streams into the generation process effectively, we design a Dual-Path Conditioned Diffusion Decoder (DPCD), which uses cross-attention for semantic conditions and ControlNet residuals for dense visual guidance. Extensive experiments demonstrate that SDGIC improves semantic consistency at ultra-low bitrates while maintaining favorable perceptual quality, with a 23.4% reduction in AFINE on the CLIC2020 dataset.