Abstract

Distributed source coding is proposed as a new video compression paradigm to shift the bulk of computation to the decoder, in order to achieve low complexity encoder design. During the decoding operation, the decoder will iteratively request for syndrome bits from the encoder to achieve certain decoding quality. This causes inconsistent bitrate and might increase the bitrate. In this study, we add a module at the decoder to iteratively predict the next to be decoded frame. Hence, the bitrate can be controlled by eliminating the need to request for more syndrome bits from the encoder. The results show that the proposed model can achieve performance close to the conventional distributed source coding model.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call