Temporal Error Concealment Algorithm Research Articles

An effective temporal error concealment in H.264 video sequences based on scene change detection-PCA model

This paper proposes a new temporal error concealment algorithm in H.264 video sequences based on scene change detection and PCA model. In order to detect scene change, dynamic threshold and image similarity metric are presented using coding prediction mode and DCT AC energy in H.264 baseline. UPCA (Updated PCA) model is presented by combining the scene change feature with Index transformation-Buffer updating approach. The lost images are concealed by Projection onto Convex Sets algorithm with UPCA model. Experimental results show that the proposed algorithm can achieve better error concealment performance for the higher motion and the frequent scene change, compared with the related method.

H.264/AVC video error concealment algorithm by employing motion vector recovery under cloud computing environment

Network-based cloud computing has rapidly expanded as an effective way of video processing and transmission. Since packet losses or errors may frequently occur in cloud computing environment during the transmission of compressed video, error concealment is applied in the decoder to prevent significant degradation of image quality. Motion vector (MV) recovery is a widely-used temporal error concealment which shows satisfactory performance in practical application of video transmission. In this paper, a fast and effective temporal error concealment algorithm for H.264/AVC is presented, which efficiently utilizes the MVs of neighboring macroblocks (MB) which are adjacent to the lost MB under different circumstances. To ensure the precision of the MV recovery, a smallest division of $$4\times 4$$ 4 × 4 sub-block is applied, which will not bring too much complexity in the proposed algorithm. Each MV of sub-block is restored in individual method, and the recovery information is gathered from the nearby 20 sub-blocks. Simulation results under the virtual cloud environment show that our scheme can highly improve the quality of reconstructed video and obtain a gain of about 4 dB in PSNR, compared with other temporal error concealment methods in the condition of different packet loss rates and quantization parameters. The practical simplicity ensures that the proposed method can be readily applied to real-time video applications running under cloud computing environment.

The switching-activity estimation of SAD engine for video coding systems

The sum of absolute difference (SAD) is generally adopted as a cost function in motion estimation (ME) and temporal error concealment (TEC) algorithm owing to its efficiency. The hardware architecture of SAD also consumes considerable power dissipation in video codec chip. Hence, the switching-activity analysis on SAD is quite essential from algorithm/architecture perspectives. This work develops the estimation formulas for switching-activity dedicated for SAD engine according to probability theory. The experiment results reveal that the probability error rate (PER) of the SAD engine is as minor as 5.61%. Consequently, this leads to a precise switching-activity estimation of the SAD-based algorithm/architecture for video signal processing.

The successively temporal error concealment algorithm using error-adaptive block matching principle

Generally, the temporal error concealment (TEC) adopts the blocks around the corrupted block (CB) as the search pattern to find the best-match block in previous frame. Once the CB is recovered, it is referred to as the recovered block (RB). Although RB can be the search pattern to find the best-match block of another CB, RB is not the same as its original block (OB). The error between the RB and its OB limits the performance of TEC. The successively temporal error concealment (STEC) algorithm is proposed to alleviate this error. The STEC procedure consists of tier-1 and tier-2. The tier-1 divides a corrupted macroblock into four corrupted 8 × 8 blocks and generates a recovering order for them. The corrupted 8 × 8 block with the first place of recovering order is recovered in tier-1, and remaining 8 × 8 CBs are recovered in tier-2 along the recovering order. In tier-2, the error-adaptive block matching principle (EA-BMP) is proposed for the RB as the search pattern to recover remaining corrupted 8 × 8 blocks. The proposed STEC outperforms sophisticated TEC algorithms on average PSNR by 0.3 dB on the packet error rate of 20% at least.

Temporal Error Concealment Algorithm with Improved Ability on Motion Vector Recovery for H.264/AVC Decoder

During network transmissions, packet loss cannot be avoided, which leads to the decrease of the video quality. In order to provide better quality under unreliable communication environments, an efficient motion vector (MV) recovery algorithm is proposed in this paper for video temporal error concealment (TEC) with H.264/AVC standard. To increase the accuracy, 4 × 4 block size is regarded as the basic operation unit in the proposal, and the detailed processing procedures of the proposed algorithm for three cases of packet loss which happen frequently are separately discussed. Simulation results show that the proposed algorithm could achieve better subjective and objective qualities with less time-cost compared with the existing temporal EC algorithms. With its simple composition, it is could be easily used in multimedia communication systems with H.264/AVC coding standard.

ContentAdaptive Refined Error Concealment Schemes for H.264/AVC Video Coding

this paper, we propose content-adaptive refined error concealment schemes for video transmission over packet lossy networks. The proposed spatial and temporal error concealment algorithms improve the reconstructed video quality at the common loss rates encountered on the networks. To reduce the computational complexity, the proposed schemes make use of the existing information of Intra prediction modes from the coded bit-stream for spatial error concealment. Neighboring macroblocks mode information is used to adapt the partition size of the lost macroblock to enhance the temporal error concealment performance. In addition, overlapped motion compensation is used to avoid the spatial discontinuities. A hybrid spatio-temporal concealment scheme is also proposed to use the appropriate algorithm for the lost macroblocks in the Inter frames. Experimental results demonstrate that the proposed schemes improve the concealment performance and reduce the structural degradations.

Efficient temporal error concealment algorithm for H.264/AVC inter frame decoding

SUMMARYDuring the real‐time video transmission over error‐prone network, the compressed video signal is very sensitive to the channel disturbances, and the packet loss cannot be avoided which causes the video quality reduction in the destination. To satisfy the audiences' visual feeling, the decoder attempts to conceal the error effect by using the spatial or temporal domain information to estimate the missing video content. In this paper, an efficient temporal error concealment (ETEC) algorithm is proposed for H.264/AVC Inter frame decoding to eliminate the error effect for Human Visual System (HVS). The 4 × 4 block size is used as the basic motion vector (MV) recovery unit to increase the error concealment (EC) accuracy, and the MV of the lost macroblock (MB) is recovered by employing the MV information of the neighboring intact MBs based on the geometry and data interpolation. The simulation results show that the proposed algorithm can achieve better performance compared with the existing TEC methods not only for the objective quality measurement, but also for the HVS subjective view perception. Because of its simple composition, the proposed algorithm is pervasive to be used in the real‐time multimedia communication systems with the video coding standard H.264/AVC. Copyright © 2011 John Wiley & Sons, Ltd.

Temporal error concealment algorithm designed for H.264

To overcome the degradation of video quality caused by transmission errors,a new temporal error concealment algorithm based on motion vector adaptive recovery was proposed,which efficiently exploited the characteristic of H.264.Firstly the Motion Vector Relative Strength(MVRS) was estimated by calculating a ratio involving the motion vectors of adjacent macro blocks.Then either minimum variance of boundary motion vectors method or polynomial interpolation method was employed to recover the lost motion vectors according to the magnitude of MVRS.The simulation results show that the proposed algorithm can highly improve the quality of reconstructed video and obtain a gain of about 0.2~2.5dB in PSNR,compared with conventional temporal error concealment algorithms in the condition of equal packet loss rate.

Temporal Error Concealment for MPEG-Coded Video over Error-Prone Networks

Compressed video streams that transmitted over error prone networks may suffer severe degradation. Error concealment (EC) is one practical approach to combat with the error propagation. A number of EC techniques have been developed to diminish the influence of transmission errors. However, the techniques are always inefficient when the motions of object in a video are fast or complex. This paper proposes an adaptive temporal EC algorithm to precisely conceal the errors for MPEG-coded video. In the proposed method, each damaged macroblock is first divided into four equal sized sub-blocks. Both of the undamaged motion vectors and gray intensity information surrounding a damaged macroblock are performed to reconstruct the lost motion vectors of the sub-blocks based on boundary matching technique. Then the estimated motion vectors are utilized to reconstruct the damaged macroblock by exploiting the information from reference frames. Experimental results show that the visual quality and the PSNR evaluation of reconstructed frames are significantly improved by using the proposed EC algorithms. Thus the proposed method is expected to be a useful EC algorithm for motion vector compressed video over an error-prone network.

Combined and Iterative Form of Spatial and Temporal Error Concealment for Video Signals

Video transmission over unreliable channels may suffer from the data loss or corruption. To facilitate the transmission, video content is compressed to save the bandwidth. The compression algorithms remove the redundancies in the video signal, meanwhile increasing the dependencies among symbols in the compressed bit-stream. Thus, when errors occur, they may propagate in both space and time. Among various error control techniques, error concealment (EC) is an effective method that is performed at the decoder to mitigate the influence of errors on the quality of reconstructed images. In this paper, a joint spatial and temporal EC algorithm (JSTEC) is presented. First, several existing temporal EC algorithms are combined in an appropriate order. Then, the spatial correlation in the video is exploited in an iterative form to improve the performance of the temporal EC. Experimental results show that JSTEC performs better both in peak signal-to-noise ratio and subjective quality of images than the existing algorithms

Temporal Error Concealment Based on Weighted Pixel Estimator

A temporal error concealment algorithm for the block-based video coder has been proposed. The concept of block overlapping is adopted to conceal the erroneous blocks, and the recovered pixels are estimated by the weighted sum from the overlapping. The overlapping weighting matrix has been carefully selected in order to fully exploit the spatial-temporal correlation between boundary blocks and the lost block. Furthermore, the motion vector for the lost block has been recovered by considering the best results for the overlapping. The experimental results are shown by integrating our algorithm into the H.263+ coder.

Temporal Error Concealment for H.264 Video Based on Adaptive Block-Size Pixel Replacement

The H.264 standard allows each macroblock to have up to sixteen motion vectors, four reference frames, and a macroblock mode. Exploiting this feature, we present an efficient temporal error concealment algorithm for H.264-coded video. The proposed method turns out to show good performance compared with conventional approaches.

Temporal error concealment algorithm by recursive block-matching principle

Compressed video bitstream is sensitive to channel errors that may degrade the reconstructed pictures severely. Error concealment is an effective approach to reduce the impact of errors, and it is implemented at the decoders. Its aim is to mask the effect of the missing blocks to create image or video of subjectively acceptable quality by using the spatial or/and temporal correlation. This paper proposes a temporal error concealment method that exploits the temporal redundancy of a video sequence using a block-matching approach. A recursive block-matching algorithm, based on the two-step block-matching principle, is presented to refine the concealed video. The proposed algorithm can also conceal the multiple erroneous slices. The proposed algorithm outperforms the algorithms presented in previous studies.

Novel Error Concealment Method With Adaptive Prediction to the Abrupt and Gradual Scene Changes

In this paper, the impact of scene change on the conventional error concealment method is addressed and a novel error concealment method is proposed to improve the insufficiency of the conventional temporal error concealment algorithm due to the occurrence of scene change. Combining with the low complexity scene change detection algorithm using macroblock type information, the corrupt blocks resulting from bit errors are concealed either temporally or spatially depending on whether or not an abrupt scene change is found. In the case of gradual scene change, a novel error concealment method of interpolation and extrapolation is proposed to utilize the linear property of gradual scene change sequence, and effectively reduce the concealment error in comparison with the conventional algorithm. A great improvement of about 3 to 5 dB PSNR on average and 6 to 8 dB in some cases is obtained with very little memory and computation overhead.