Search

Resilience to channel noise is indispensable for video communication based consumer applications. Temporal error concealment is a kind of decoder-based technique to compensate for transmission errors and is popular in most consumer electronics. In this paper, an adaptive temporal error concealment scheme is proposed based on H.264 video standard to improve error resilience ability for video consumer applications. Mode switching mechanism is devised to flexibly choose appropriate temporal concealment mode. The search range for candidate motion vectors could be adaptively determined by the range of spatially and temporally neighboring motion vectors. Weighted outer boundary distortion function could dynamically select the optimal replacing motion vector for local area. Experiment and comparison with recent error concealment schemes shows that our proposed algorithm could effectively improve received video quality with reduced requirement for decoding time. Index Terms--Error concealment, video, temporal, motion vector.

Error resilience is an essential problem for video communications, such as digital TV broadcasting, mobile video terminals and video telephone. The latest video compression standard H.264/AVC provides more coding efficiency for a wide range of video consumer applications. Yet H.264 video streams are still vulnerable to transmission errors. In this paper, a set of error concealment techniques are proposed to provide error resilience based on new coding and network characteristics of H.264. The temporal concealment involves a method of subblock-based refined motion compensated concealment using weighting boundary match, which improves the ability to deal with high motion activity areas. The spatial concealment scheme involves an algorithm of refined directional weighted spatial interpolation, which could protect object edge integrity. Combining the above algorithms, an adaptive spatial/temporal estimation method with low complexity is presented. Transmission over typical 3GPP/3GPP2 mobile IP channels is simulated with a wide range of bit rate and BER. The refined concealment techniques provide more error robustness for video consumer electronics than those suggested in H.264 without any encoder-side modifications (1). Index Terms--Error concealment, H.264 video, DTV, mobile IP, resilience.

The new H.264 video coding standard has a significant performance benefit versus older standards in compression performance. Unfortunately, it also has a considerable increase in encoder complexity. In this paper, we extend and adapt the concept of detection of all-zero quantization coefficients before transformation and quantization. A guideline for miss detection and false detection is proposed according to the principle of two-dimensional integer DCT and quantization in H.264. With this guideline, a generalized discriminant that can be applied to different integer transform is also provided. It is so-called early detection method of all-zero Integer DCT coefficients (EDMAIDC). Simulation results demonstrate the efficiency of proposed algorithm of reducing encoding complexity while incurring no lass in quality (1). Index Terms--integer transform, quantization, all-zero coefficient block

This paper proposes an efficient hardware architecture to accelerate adaptive deblocking filter of H.264/A VC video coding. Compact data access unit, Line-of-Pixel (LOP) is defined in this paper. Line-of-Pixel and building data buffer are employed to simplify data exchange between deblocking filter and outside data memory. Edge filter, which is designed to process each group of pixels on both sides of one edge, is kernel of deblocking filter. It is implemented in multiple parallel pipelines to increase efficiency. By carefully design, the proposed deblocking filter can be embedded in general-purpose processor or DSP to support special instructions for acceleration of software codec. This filter also can be used to full hardware H.264/AVC codec (1). Index Terms--Deblocking filter, FPGA, H.264/AVC, video coding.

Seamless splicing is to concatenate two different video streams into one video stream with the Video Buffer Verifier (VBV) constraint being satisfied in this paper we thoroughly analyze the seamless splicing process of MPEG-2 video stream in compressed domain. We first analyze the selection of frame conversion schemes and present a drift-free frame conversion scheme in compressed domain. Then we acquire the target number of bit in the splicing period by the vbv_delay value of the streams to be spliced, and adopt p-domain rate control to control the bit rate. Experiment results show that the proposed algorithm can splice bitstreams efficiently with image degradation restricted in a few number of frames (1). Index Terms--Seamless Splicing, Frame Conversion, Rate Control.