Learned Distributed Image Compression with Multi-Scale Patch Matching in Feature Domain

Beyond achieving higher compression efficiency over classical image codecs, deep is expected to be improved with additional side information, e.g., another from a different perspective of the same scene. To better utilize information under distributed scenario, existing method only implements patch matching at domain solve parallax problem caused by difference in viewing points. However, not robust variance scale, shape, and illumination angles, can make full use rich texture image. resolve this issue, we propose Multi-Scale Feature Domain Patch Matching (MSFDPM) fully utilizes decoder model. Specifically, MSFDPM consists feature extractor, multi-scale module, fusion network. Furthermore, reuse inter-patch correlation shallow layer accelerate layer. Finally, find that our further improves rate about 20% compared domain.