多尺度图像表示压缩
Multi-scale Image Representation Compression
- Visual Information Lab, University of Bristol（布里斯托大学视觉信息实验室）
- Tencent Media Lab（腾讯媒体实验室）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
本文提出基于NVRC端到端压缩流程的过拟合图像编解码器MIRC，引入多尺度表示与跨阶段参数共享，在CLIC2020验证集上较VVC实现10.5% BD-rate节省，且解码预算可按需选择。
AI中文摘要：
过拟合编解码器在图像和视频压缩中展现出良好性能。对于图像压缩，Cool-chic系列模型相比与场景无关的模型表现出竞争力，解码复杂度低几个数量级，但代价是过拟合过程更长。然而，这些过拟合图像编解码器并未完全针对率失真目标进行优化：其网络权重在训练期间保持全精度，且相关量化参数在单独的训练后阶段选择；此外，其合成操作在单一尺度下进行，忽略了跨尺度冗余。本文提出MIRC，一种过拟合图像编解码器，其中包括潜变量、合成网络和熵模型在内的每个编码组件均在单一率失真目标下进行量化和熵编码，采用神经视频表示编解码器NVRC的端到端压缩流程。我们进一步引入具有跨阶段参数共享的多尺度表示，以少量传输开销提升编码效率。在CLIC2020专业验证集上，MIRC相比VVC（VTM 22.0）实现了10.5%的BD-rate节省。此外，MIRC提供了一系列配置，每像素覆盖1.2至2.9 kMAC，因此可选择解码预算以匹配部署目标。
英文摘要：
Overfitted codecs have demonstrated promising performance for image and video compression. In particular, for image compression, the Cool-chic family of models has shown competitive performance against scene-agnostic models, with orders of magnitude lower decoding complexity at the cost of a longer overfitting process. However, these overfitted image codecs are not fully optimized toward the rate-distortion objective: their network weights remain in full precision during training, and the associated quantization parameters are selected in a separate post-training stage. Furthermore, their synthesis operates at a single scale, which overlooks cross-scale redundancy. In this paper, we propose MIRC, an overfitted image codec in which every coded component, including the latents, the synthesis network, and the entropy models, is quantized and entropy coded under a single rate-distortion objective, adopting the end-to-end compression pipeline of the neural video representation codec NVRC. We further introduce a multi-scale representation with cross-stage parameter sharing, which improves coding efficiency at a small transmitted overhead. On the CLIC2020 professional validation set, MIRC achieves a 10.5% BD-rate saving against VVC (VTM 22.0). Moreover, MIRC offers a family of configurations spanning 1.2 to 2.9 kMAC per pixel, so the decoding budget can be selected to match the deployment target.
