Latent Discretization for Continuous-time Sequence Compression
This work addresses compression efficiency for domains like video and motion capture, offering a novel method that is incremental in improving existing neural compression approaches.
The paper tackles the problem of high computational cost in neural sequence compression by modeling sequences as continuous-time processes and learning efficient discretization, achieving reductions in bit rates in video and motion capture compression.
Neural compression offers a domain-agnostic approach to creating codecs for lossy or lossless compression via deep generative models. For sequence compression, however, most deep sequence models have costs that scale with the sequence length rather than the sequence complexity. In this work, we instead treat data sequences as observations from an underlying continuous-time process and learn how to efficiently discretize while retaining information about the full sequence. As a consequence of decoupling sequential information from its temporal discretization, our approach allows for greater compression rates and smaller computational complexity. Moreover, the continuous-time approach naturally allows us to decode at different time intervals. We empirically verify our approach on multiple domains involving compression of video and motion capture sequences, showing that our approaches can automatically achieve reductions in bit rates by learning how to discretize.