Key points are not available for this paper at this time.
This paper presents CompressAI, a platform that provides custom operations, , models and tools to research, develop and evaluate end-to-end image and compression codecs. In particular, CompressAI includes pre-trained models evaluation tools to compare learned methods with traditional codecs. models from the state-of-the-art on learned end-to-end compression thus been reimplemented in PyTorch and trained from scratch. We also objective comparison results using PSNR and MS-SSIM metrics vs. -rate, using the Kodak image dataset as test set. Although this framework implements models for still-picture compression, it is intended to be extended to the video compression domain.
Bégaint et al. (Thu,) studied this question.