PerCoV2：基于隐式分层掩码图像建模的改进型超低比特率感知图像压缩

PerCoV2: Improved Ultra-Low Bit-Rate Perceptual Image Compression with Implicit Hierarchical Masked Image Modeling

March 12, 2025

作者: Nikolai Körber, Eduard Kromer, Andreas Siebert, Sascha Hauke, Daniel Mueller-Gritschneder, Björn Schuller

cs.AI

摘要

我們推出PerCoV2，這是一種新穎且開放的超低比特率感知圖像壓縮系統，專為帶寬和存儲受限的應用而設計。基於Careil等人的先前工作，PerCoV2將原始框架擴展至Stable Diffusion 3生態系統，並通過顯式建模離散超潛在圖像分佈來提升熵編碼效率。為此，我們對最新的自回歸方法（VAR和MaskGIT）進行了全面的比較，並在大規模MSCOCO-30k基準上評估了我們的方法。與之前的工作相比，PerCoV2（i）在更低的比特率下實現了更高的圖像保真度，同時保持了競爭性的感知質量，（ii）引入了混合生成模式以進一步節省比特率，以及（iii）完全基於公開組件構建。代碼和訓練模型將在https://github.com/Nikolai10/PerCoV2上發布。

English

We introduce PerCoV2, a novel and open ultra-low bit-rate perceptual image compression system designed for bandwidth- and storage-constrained applications. Building upon prior work by Careil et al., PerCoV2 extends the original formulation to the Stable Diffusion 3 ecosystem and enhances entropy coding efficiency by explicitly modeling the discrete hyper-latent image distribution. To this end, we conduct a comprehensive comparison of recent autoregressive methods (VAR and MaskGIT) for entropy modeling and evaluate our approach on the large-scale MSCOCO-30k benchmark. Compared to previous work, PerCoV2 (i) achieves higher image fidelity at even lower bit-rates while maintaining competitive perceptual quality, (ii) features a hybrid generation mode for further bit-rate savings, and (iii) is built solely on public components. Code and trained models will be released at https://github.com/Nikolai10/PerCoV2.