发表机构
Beijing Institute of Computer Technology and Application; School of Elite Engineering, Northwestern Polytechnical University（北京计算机技术及应用研究所; 西北工业大学精英工程学院）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文针对学习型图像压缩模型低比特部署的效率与精度问题，提出带四阶段优化的 HAMP-LIC 框架，实现最高4.85倍模型压缩且仅0.59%率失真损失，性能优于现有方法并消除跨平台编解码误差。
AI 中文摘要
学习型图像压缩（Learned Image Compression, LIC）模型具备优异的率失真性能，但受限于异构硬件平台间的高计算复杂度及编解码不匹配问题。均匀固定精度量化可缓解上述问题，但因未考虑各层量化敏感性的差异，在低比特宽下会出现严重的质量退化。为实现预训练 LIC 模型高效且精准的低比特部署，本文提出 HAMP-LIC，这是一种带有四阶段优化策略的 Hessian 感知混合精度后训练量化（Post-Training Quantization, PTQ）框架。第一阶段，基于 Hessian 迹估计块级敏感性，以捕捉二阶重要性；第二阶段，任务感知细化模块通过联合考量量化失真与率失真性能，调整上述敏感性；第三阶段，在全局模型规模约束下，依据细化后的敏感性分布分配比特宽，以平衡效率与重构质量；第四阶段，利用小校准集进行块级重构，进一步抑制量化误差。在 Minnen2018、Cheng2020 等代表性 LIC 模型上开展的实验表明，HAMP-LIC 可实现最高达 4.85 倍的模型压缩，且仅产生 0.59% 的 BD-rate 损失；在多个数据集上，其性能始终优于现有固定精度及混合精度 PTQ 方法，同时完全消除了跨平台编解码误差。
英文摘要
Use this plain-text version for the arXiv abstract field: Learned image compression (LIC) models achieve strong rate-distortion performance but are hindered by high computational complexity and encoding-decoding mismatches across heterogeneous hardware platforms. Uniform fixed-precision quantization alleviates these issues but suffers severe quality degradation at low bit widths because it ignores differences in the quantization sensitivities of individual layers. To enable efficient and accurate low-bit deployment of pretrained LIC models, we propose HAMP-LIC, a Hessian-aware mixed-precision post-training quantization (PTQ) framework with a four-stage optimization strategy. First, block-wise sensitivity is estimated from the Hessian trace to capture second-order importance. Second, a task-aware refinement module adjusts these sensitivities by jointly considering quantization distortion and rate-distortion performance. Third, guided by the refined sensitivity profile, bit widths are allocated under a global model-size constraint to balance efficiency and reconstruction quality. Finally, block-wise reconstruction using a small calibration set further suppresses quantization error. Experiments on representative LIC models, including Minnen2018 and Cheng2020, demonstrate that HAMP-LIC achieves up to 4.85x model compression with as little as 0.59% BD-rate loss. It consistently outperforms existing fixed- and mixed-precision PTQ methods across multiple datasets while completely eliminating cross-platform encoding-decoding errors.
CommentsLearned image compression, post-training quantization, mixed-precision quantization, Hessian-based sensitivity analysis, model compression
