ProgDTD: Progressive Learned Image Compression with Double-Tail-Drop Training
This addresses the need for improved user experience in slow network conditions by making learned image compression progressive, but it is incremental as it builds on existing CNN-based models.
The paper tackles the problem of enabling progressive image compression in learned models, which allows images to load from low to high resolution, and introduces ProgDTD, a training method that transforms non-progressive models into progressive ones without adding parameters, achieving performance comparable to state-of-the-art models in terms of MS-SSIM and accuracy.
Progressive compression allows images to start loading as low-resolution versions, becoming clearer as more data is received. This increases user experience when, for example, network connections are slow. Today, most approaches for image compression, both classical and learned ones, are designed to be non-progressive. This paper introduces ProgDTD, a training method that transforms learned, non-progressive image compression approaches into progressive ones. The design of ProgDTD is based on the observation that the information stored within the bottleneck of a compression model commonly varies in importance. To create a progressive compression model, ProgDTD modifies the training steps to enforce the model to store the data in the bottleneck sorted by priority. We achieve progressive compression by transmitting the data in order of its sorted index. ProgDTD is designed for CNN-based learned image compression models, does not need additional parameters, and has a customizable range of progressiveness. For evaluation, we apply ProgDTDto the hyperprior model, one of the most common structures in learned image compression. Our experimental results show that ProgDTD performs comparably to its non-progressive counterparts and other state-of-the-art progressive models in terms of MS-SSIM and accuracy.