In Learned Image Compression (LIC), a model is trained at encoding and decoding images sampled from a source domain, often outperforming traditional codecs on natural images; yet its performance may be far from optimal on images sampled from different domains. In this work, we tackle the problem of adapting a pre-trained model to multiple target domains by plugging into the decoder an adapter module for each of them, including the source one. Each adapter improves the decoder performance on a specific domain, without the model forgetting about the images seen at training time. A gate network computes the weights to optimally blend the contributions from the adapters when the bitstream is decoded. We experimentally validate our method over two state-of-the-art pre-trained models, observing improved rate-distortion efficiency on the target domains without penalties on the source domain. Furthermore, the gate's ability to find similarities with the learned target domains enables better encoding efficiency also for images outside them.

翻译：在图像压缩学习中，模型通过编码和解码从源域采样的图像进行训练，通常在自然图像上的性能优于传统编解码器；然而，对于来自不同域的图像，其性能可能远非最优。在本工作中，我们通过为每个目标域（包括源域）在解码器中插入适配器模块，来解决将预训练模型适配到多个目标域的问题。每个适配器在不遗忘训练时所见图像的前提下，提升解码器在特定域的性能。网络门控机制可计算权重，以在比特流解码时最优融合各适配器的贡献。我们通过两个最先进的预训练模型进行实验验证，观察到在目标域上率失真效率提升，且对源域无性能损失。此外，门控机制能识别与已学习目标域的相似性，从而对域外图像也能实现更高的编码效率。