Detecting corrupted intra macroblocks in H.263 video

EUROCON 2007 - The International Conference on "Computer as a Tool"

2007

The video compression standards commonly adopted in wireless multimedia services utilize variable length codes (VLC) in order to attain high compression ratios. While providing the high data rates required, this technique makes the system more susceptible to transmission errors. Thus the end-to-end quality of the video stream transmitted over an error-prone channel depends on the detection, and concealment of the corrupted macroblocks. The error detection capability of standard decoders is quite limited, for example, in the case of the H.263+ codec around 40.54% of the corrupted macroblocks are undetected, placing a bound on the perceived quality of the reconstructed video sequence. This paper presents a novel solution using eight pixel domain dissimilarity metrics computed in the CIE LUV color space which can be used at decode time to improve the error detection rate of the standard decoder. The spatial dissimilarity metric has been found to perform the best with an average increase in error detection rate of 60.38% when compared to the standard decoder (about 20% more than other published results) with 0% of false detection and a gain in peak signal-to-noise ratio (PSNR) of 3.94 dB.

Section: Introductionmentioning

confidence: 99%

Enhancing the Error Detection Capabilities of the Standard Video Decoder using Pixel Domain Dissimilarity Metrics

Farrugia

EUROCON 2007 - The International Conference on "Computer as a Tool"

2007

“…However, this method only manages to detect 57% of the corrupted MBs. Different error detection and concealment methods have been proposed which exploit the inherent redundancies within and outside an MB [5] - [7] at pixel level. However, these features achieve low detection rates when applied individually.…”

Section: Introductionmentioning

confidence: 99%

Robust transmission of H.264/AVC sequences using list decoding and source constraints

Farrugia¹,

MELECON 2008 - The 14th IEEE Mediterranean Electrotechnical Conference

2008

Abstract-Video compression standards commonly adopted in wireless multimedia transmission systems generally apply variable length codes (VLCs) to achieve the required datarates. The resulting bitstreams are sensitive to transmission errors which cause visual artifacts that propagate in both spatial and temporal domains, thus reducing significantly the quality of the reconstructed video sequences. This paper proposes the application of a List Decoder to improve the quality of H.264/AVC encoded sequences. This method relies on sequence estimation and residual source redundancies to recover feasible images. The recovered images may still contain errors which may cause visually distorted macroblocks (MBs). Therefore, the Average Intersample Difference across Boundaries (AIDB) metric was used to detect these visually impaired MBs at pixel level which are then adequately concealed. Simulation results show that the proposed algorithm manages to achieve Peak Signal-to-Noise (PSNR) gains of up to 2.17 dB when compared to the conventional decoding method. The proposed solution is fully compatible with the H.264/AVC standard and no additional bandwidth is required. Index Terms-Error correction coding, error resilient coding, H.264/AVC, video coding, wireless video transmission. I. INTRODUCTIONULTIMEDIA communication services necessitate efficient video compression standards in order to deliver high quality content within limited bandwidths. H.264/AVC seems to be an attractive candidate for all wireless video applications mainly because of its enhanced compression efficiency and network friendly design [1]. However, it is sensitive to transmission errors, which are common in wireless environments, where a single corrupted bit may cause visual distortions which will significantly reduce the perceptual quality of the reconstructed video sequence.A feedback channel, commonly used for reliable data transmission, cannot be adopted in typical conversational and multicast/broadcast applications because of delay constraints and the unavailability of the feedback channel respectively. Moreover, low bit-rate and low delay applications generally adopt the baseline profile which has limited error-resilient mechanisms [2], [3].In order to solve these problems, a number of errorresilient mechanisms were proposed in literature. authors have applied syntax analysis in order to detect corrupted macroblocks (MBs). However, this method only manages to detect 57% of the corrupted MBs. Different error detection and concealment methods have been proposed which exploit the inherent redundancies within and outside an MB [5] -[7] at pixel level. However, these features achieve low detection rates when applied individually. To alleviate this problem, the authors in [8] have combined a set of pixel level features and applied heuristic thresholds. They have also applied an iterative error correction algorithm which has achieved a significant gain in Peak Signal-to-Noise Ratio (PSNR) at the expense of a significant increase in complexity, which makes ...

“…A number of pixel-level error detection algorithms are present in literature [6], [12] - [17]. However, these algorithms only manage to detect 20% -40% of the corrupted macroblocks.…”

Section: Introductionmentioning

confidence: 99%

Enhancing the Error Detection Capabilities of DCT Based Codecs using Compressed Domain Dissimilarity Metrics

Farrugia

EUROCON 2007 - The International Conference on "Computer as a Tool"

2007

Video compression standards are implemented in wireless data transmission technologies to provide multimedia services efficiently. These compression standards generally utilize the Discrete Cosine Transform (DCT) in conjunction with variable length codes (VLC) in order to achieve the required high compression ratios. While providing the necessary high data rates, this technique has the disadvantage of making the system more susceptible to transmission errors. The standard decoders do not manage to detect a large number of corrupted macroblocks, 40.54% not detected for H.263+, contributing to a significant reduction in the end-to-end video quality as perceived by the enduser. This paper presents three dissimilarity metrics which contain both color and texture information and that can be extracted directly from the compressed DCT coefficients. These metrics can be used to enhance the error-detection capabilities of standard DCT based codecs. Simulation results show that the proposed algorithm increases the error detection rate by 54.06% with a gain in peak signal-to-noise ratio (PSNR) of 3.21 dB. This improvement in performance is superior to other solutions found in literature.