Abstract:In existing video coding schemes with spatial scalability based on pyramid frame representation, such as the ongoing H.264/MPEG-4 SVC (scalable video coding) standard, video frame at a high resolution is mainly predicted either from the lower-resolution image of the same frame or from the temporal neighboring frames at the same resolution. Most of these prediction techniques fail to exploit the two correlations simultaneously and efficiently. This paper extends the in-scale prediction technique developed for wavelet video coding to a generalized in-scale motion compensation framework for H.264/MPEG-4 SVC. In this framework, for a video frame at a high resolution layer, the lowpass content is predicted from the information already coded in lower resolution layer, but the highpass content is predicted by exploiting the neighboring frames at current resolution. In this way, both the cross-resolution correlation and temporal correlation are exploited simultaneously, which leads to much higher efficiency in prediction. Preliminary experimental results demonstrate that the proposed framework improves the spatial scalability performance of current H.264/MPEG-4 SVC. The improvement is significant especially for high-fidelity video coding. In addition, another advantage over wavelet-based in-scale scheme is achieved that the proposed framework can support arbitrary down-sampling and up-sampling filters.

A novel spatial-temporal position prediction motion-compensated interpolation for frame rate up-conversion

Position Prediction Motion-Compensated Interpolation For Frame Rate Up-Conversion Using Temporal Modeling

Visual-weighted Motion Compensation Frame Interpolation with Motion Vector Refinement

Bi-Direction Motion Vector Retrieval Based Error Concealment Scheme for H.264/Avc

Selective VS-MRF-ME and intra coding in H.264 based on spatiotemporal continuity of motion field

Efficient interpolation architecture design for motion compensation in H.264/AVC

Efficient Motion Vector Interpolation for Error Concealment of H.264/AVC

Multi-View Video Coding Based on Vector Estimation and Weighted Disparity Interpolation

Multi-view Video Coding Based on View Prediction

Video Frame Interpolation Using 3-D Total Variation Regularized Completion

A new frame interpolation method with pixel-level motion vector field

An Adaptive Frame Interpolation Algorithm Using Statistic Analysis Of Motions And Residual Energy

Generalized In-Scale Motion Compensation Framework for Spatial Scalable Video Coding.

An Efficient 3D Video Frame Interpolation Method Using Color-Depth-motion Information

Frame Rate Up Conversion Via Bayesian Motion Estimation

Adaptive Motion Vector Prediction Based On Spatiotemporal Correlation

Fast multi-frame based motion estimation algorithm for MPEG-4 to H.264 video transcoder

Adaptive motion-compensated interpolation based on spatiotemporal segmentation

An Interpolation Method by Predicting the Direction of Pixel Texture Changing Trend for H.264/AVC Intra Prediction

Low-Cost Multi-Hypothesis Motion Compensation For Video Coding