Learned layered coding for Successive Refinement in the Wyner-Ziv Problem
This work addresses the Wyner-Ziv problem for scalable coding with side information, offering an incremental improvement through a learned approach.
The paper tackled the problem of successively refining a continuous source with correlated side information in the Wyner-Ziv coding setup by proposing a data-driven approach using recurrent neural networks (RNNs) to learn layered encoders and decoders, achieving rate-distortion performance on par with monolithic Wyner-Ziv coding and close to the theoretical bound.
We propose a data-driven approach to explicitly learn the progressive encoding of a continuous source, which is successively decoded with increasing levels of quality and with the aid of correlated side information. This setup refers to the successive refinement of the Wyner-Ziv coding problem. Assuming ideal Slepian-Wolf coding, our approach employs recurrent neural networks (RNNs) to learn layered encoders and decoders for the quadratic Gaussian case. The models are trained by minimizing a variational bound on the rate-distortion function of the successively refined Wyner-Ziv coding problem. We demonstrate that RNNs can explicitly retrieve layered binning solutions akin to scalable nested quantization. Moreover, the rate-distortion performance of the scheme is on par with the corresponding monolithic Wyner-Ziv coding approach and is close to the rate-distortion bound.