Applying data mining techniques in a Wyner-Ziv to H.264 video transcoder
IWANN'11 Proceedings of the 11th international conference on Artificial neural networks conference on Advances in computational intelligence - Volume Part I
Scalable mobile-to-mobile video communications based on an improved WZ-to-SVC transcoder
MMM'12 Proceedings of the 18th international conference on Advances in Multimedia Modeling
Fast mode decision algorithm for H.264/AVC-to-SVC transcoding with temporal scalability
MMM'12 Proceedings of the 18th international conference on Advances in Multimedia Modeling
| Hi-index | 0.43 |
This paper proposes a Wyner-Ziv/H.264 transcoder that enables low cost video applications. The proposed solution supports video encoding on resource constrained devices such as disposable video cameras, network camcorders and low cost video encoders. This approach is based on reducing encoding resource requirements on a device by using Wyner-Ziv video encoding. The system shifts the burden of complexity away from the encoder, for example to a network node, where a transcoder efficiently converts WZ encoded video to H.264 by reusing the information from the WZ decoding stage. The transcoded H.264 video is requires fewer resources than WZ decoding and therefore reduces the complexity of decoding. The complexity of encoding and playback ends of video applications is thus reduced enabling new class of consumer application. The paper is focused on reducing the complexity of the macro-block mode coding decision process carried out in H.264 encoding stage of the transcoder. Based on a data mining process, the approach replaces the high complexity H.264 mode decision algorithm by a faster decision tree. The proposed architecture reduces the battery consumption of the end-user devices and the transcoding time is reduced by 86% with negligible rate-distortion loss.