Video transcoder architectures for bit rate scaling of H.263 bit streams

Abstract

Video transcoding is one of the key technologies in implementing dynamic adaptation of the bit rate of a pre-encoded video stream to the available bandwidth over various networks. Many different transcoder architectures have been proposed to achieve fast processing. However, they suffer from quality degradation due to the drift error. In this paper, we investigate the drift caused by the fast transcoder architectures for transcoding H.263 bitstreams. We also discuss the limitations of the existing fast transcoder architectures and the flexibility that can be offered by a cascaded pixel-domain transcoder. In favor of drift-free performance and transcoding flexibility, we propose methods to reduce the computational complexity of the cascaded pixel-domain transcoder.

Digital Video Transcoding

“…Generally, there exist three transcoding architectures for the bit-rate transcoding: open-loop transcoders [23]- [25], cascaded pixel-domain transcoders (CPDTs) [24], [26] and DCT-domain transcoders (DDTs) [27]- [29]. The open-loop architectures include selective transmission [23], [24], where the high-frequency DCT coefficients are discarded, and requantization [24], [25], where the DCT coefficients are requantized.…”

“…The differences accumulate and cause the video quality to deteriorate with time until an intrapicture is reached. The error accumulation caused by the encoder/decoder predictor mismatch is called drift and it may cause severe degradation to the video quality [26], [30]. It should be noted that in the following discussions, many transcoder architectures are not strictly drift free.…”

“…Thus, the motion estimation, which usually accounts for 60%-70% of the encoder computation [31], is omitted. To address another major source of computational complexity, DCT, in [26], the end-of-block locations of the output video are predicted using those of the input video, and only partial DCT is performed.…”

Motion based object tracking in MPEG-2 stream for perceptual region discriminating rate transcoding

“…Current video transcoding techniques are based on requantization [8,13]. However, requantization do not seem to be capable enough to down scale a video to the very low bit-rate required by current the Internet scenario.…”

Resource adaptive netcentric systems on active network: a self-organizing video stream that auto morphs itself while in transit via a quasi-active network

“…It adapts with respect to two critical network resources-(i) bandwidth at various links and (ii) the processing resource at the junction nodes. Though not in this form, but adaptation with respect to bandwidth has been discussed in a few other works [1,11]. However, little attention has been paid to the adaptation with respect to node capacity.…”