Motion estimation methods for overlapped block motion compensation

An extension of conventional block motion compensation (BMC), overlapped block motion compensation (OBMC) has been shown to reduce residual errors and blocking effects in motion-compensated video. However, the overlap creates a noncausal spatial dependence between blocks and complicates motion estimation (ME) for OBMC. Iterative methods have traditionally been employed for overlapped block motion estimation (OBME). For compression, the rate for the motion vector field (MVF) may also be constrained. This work considers several rate-constrained OBME algorithms, both iterative and noniterative. Experiments demonstrate that a simple raster-scan algorithm is effective as a suboptimal, noniterative solution, with comparable or better rate-distortion performance and computational complexity than iterative OBME algorithms. Depending on the application, either this method or a simple block-matching algorithm plus iteration are the most attractive of the tested OBME schemes.     

Sub-pixel motion estimation using kernel methods

Modern video codecs such as MPEG2, MPEG4-ASP and H.264 depend on sub-pixel motion estimation to optimise rate-distortion efficiency. Sub-pixel motion estimation is implemented within these standards using interpolated values at 1/2 or 1/4 pixel accuracy. By using these interpolated values, the residual energy for each predicted macroblock is reduced. However this leads to a significant increase in complexity at the encoder, especially for H.264, where the cost of an exhaustive set of macroblock segmentations needs to be estimated for optimal mode selection. This paper presents a novel scheme for sub-pixel motion estimation based on the whole-pixel SAD distribution. Both half-pixel and quarter-pixel searches are guided by a model-free estimation of the SAD surface using a two dimensional kernel method. While giving an equivalent rate distortion performance, this approach approximately halves the number of quarter-pixel search positions giving an overall speed up of approximately 10% compared to the EPZS quarter-pixel method (the state of the art H.264 optimised sub-pixel motion estimator).     

Fast motion estimation using bidirectional gradient methods

Gradient-based motion estimation methods (GMs) are considered to be in the heart of state-of-the-art registration algorithms, being able to account for both pixel and subpixel registration and to handle various motion models (translation, rotation, affine, and projective). These methods estimate the motion between two images based on the local changes in the image intensities while assuming image smoothness. This paper offers two main contributions. The first is enhancement of the GM technique by introducing two new bidirectional formulations of the GM. These improve the convergence properties for large motions. The second is that we present an analytical convergence analysis of the GM and its properties. Experimental results demonstrate the applicability of these algorithms to real images.     

Scalable motion vector coding

Scalable wavelet-based video codecs using spatial-domain motion compensated temporal filtering require a quality-scalable motion vector codec to support a large range of bit rates with optimal compression efficiency. Introduced is a new prediction-based architecture for quality-scalable motion vector coding that outperforms the state-of-the-art wavelet-based techniques previously proposed in the literature.     

Modified CLT-domain motion estimation based on phase correlation

In this paper, a Modified Complex Lapped Transform domain Motion Estimation (MCLT-ME) method to estimate the motion of video sequence is proposed. The proposed method is based on phase correlation of current window region and corresponding window region, and provides a significant improvement in motion estimation.     

Level-set-based motion estimation algorithm for multiple reference frame motion estimation

Motion estimation (ME) has a variety of applications in image processing, pattern recognition, target tracking, and video compression. In modern video compression standards such as H.264/AVC and HEVC, multiple reference frame ME (MRFME) is adopted to reduce the temporal redundancy between successive frames in a video sequence. In MRFME, the motion search process is conducted using additional reference frames, thereby obtaining better prediction signal as compared to single reference frame ME (SRFME). However, its high computational complexity makes it difficult to be utilized in real-world applications. In order to reduce the computational complexity of MRFME, this paper proposes a level-set-based ME algorithm (LSME) without any penalty in the rate-distortion (RD) performance. First, the proposed algorithm partitions the motion search space into multiple level sets based on a rate constraint. The proposed algorithm then controls the ME process on the basis of the predetermined level sets. Experimental results show that the proposed algorithm reduces the ME time by up to 83.46% as compared to the conventional full search (FS) algorithm.     

A comparison framework for breathing motion estimation methods from 4-D imaging

Motion estimation is an important issue in radiation therapy of moving organs. In particular, motion estimates from 4-D imaging can be used to compute the distribution of an absorbed dose during the therapeutic irradiation. We propose a strategy and criteria incorporating spatiotemporal information to evaluate the accuracy of model-based methods capturing breathing motion from 4-D CT images. This evaluation relies on the identification and tracking of landmarks on the 4-D CT images by medical experts. Three different experts selected more than 500 landmarks within 4-D CT images of lungs for three patients. Landmark tracking was performed at four instants of the expiration phase. Two metrics are proposed to evaluate the tracking performance of motion-estimation models. The first metric cumulates over the four instants the errors on landmark location. The second metric integrates the error over a time interval according to an a priori breathing model for the landmark spatiotemporal trajectory. This latter metric better takes into account the dynamics of the motion. A second aim of this paper is to estimate the impact of considering several phases of the respiratory cycle as compared to using only the extreme phases (end-inspiration and end-expiration). The accuracy of three motion estimation models (two image registration-based methods and a biomechanical method) is compared through the proposed metrics and statistical tools. This paper points out the interest of taking into account more frames for reliably tracking the respiratory motion.     

2种多通道SAR/GMTI相位误差估计方法

针对多通道合成孔径雷达(SAR)和地面运动目标检测(GMTI)系统,提出了2种通道相位误差估计方法:a)利用杂波的信号特征向量与其导向矢量的线性关系直接进行通道误差估计;b)通过对回波数据进行方位重采样,然后利用杂波信号特征向量张成的空间(即信号子空间)与真实导向矢量张成的空间相同的原理进行误差估计。实验证明,2种方法均能有效地进行通道相位误差估计,并且方法 b)具有更高的估计精确度。     

基于运动矢量场的双迭代全局运动估计方法

提出了一种基于运动矢量场的双迭代全局运动估计方法，该方法用两个最小二乘迭代过程来去除局部运动区域对全局运动估计的干扰。第一个迭代使用一个递减的百分比阈值来排除局部运动区域；第二个迭代过程使用一个固定的绝对阈值来检测完整的伞局运动区域并估计运动模型参数。实验结果表明双迭代法进行伞局运动估计的结果更加准确和稳定。     

Two-dimensional matched filtering for motion estimation

In this work, we describe a frequency domain technique for the estimation of multiple superimposed motions in an image sequence. The least-squares optimum approach involves the computation of the three-dimensional (3-D) Fourier transform of the sequence, followed by the detection of one or more planes in this domain with high energy concentration. We present a more efficient algorithm, based on the properties of the Radon transform and the two-dimensional (2-D) fast Fourier transform, which can sacrifice little performance for significant computational savings. We accomplish the motion detection and estimation by designing appropriate matched filters. The performance is demonstrated on two image sequences.     

Adapted differentiators for image motion estimation

The author describes the design and use of adapted differentiators for improving the accuracy of differential-based optical flow algorithms     

Approximated densities for block-based motion estimation

In block-based motion estimation (BME), the most common procedure applied is block-matching. Some applications of Kalman filtering to motion estimation have also been found, but its applicability is rather limited due to discontinuities in the dynamic behaviour. The concept of filtering by approximated densities applied to interframe BME involving exponential distributions is found by the authors to result in superior performance     

Adaptive thresholding for motion estimation prejudgement

The authors propose an adaptive method of detecting zero motion blocks for motion estimation prejudgement. By combining our method with motion estimation algorithms, much computational effort is saved while almost the same matching performance is achieved     

Successive elimination algorithm for motion estimation

The correspondence presents a fast exhaustive search algorithm for motion estimation. The basic idea is to obtain the best estimate of the motion vectors by successively eliminating the search positions in the search window and thus decreasing the number of matching evaluations that require very intensive computations. Simulation results demonstrate that although the performance of the proposed algorithm is the same as that using the exhaustive search, the computation time has been reduced significantly     

DCT-based motion estimation

We propose novel discrete cosine transform (DCT) pseudophase techniques to estimate shift/delay between two one-dimensional(1-D) signals directly from their DCT coefficients by computing the pseudophase shift hidden in DCT and then employing the sinusoidal orthogonal principles, applicable to signal delay estimation remote sensing. Under the two-dimensional (2-D) translational motion model, we further extend the pseudophase techniques to the DCT-based motion estimation (DXT-ME) algorithm for 2-D signals/images. The DXT-ME algorithm has certain advantages over the commonly used full search block-matching approach (BKM-ME) for application to video coding despite certain limitations. In addition to its robustness in a noisy environment and low computational complexity, O(M(2)) for an MxM search range in comparison to the O(N(2).M(2)) complexity of BKM-ME for an NxN block, its ability to estimate motion completely in DCT domain makes possible the fully DCT-based motion-compensated video coder structure, which has only one major component in the feedback loop instead of three as in the conventional hybrid video coder design, and thus results in a higher system throughput. Furthermore, combination of the DCT and motion estimation units can provide space for further optimization of the overall coder. In addition, the DXT-ME algorithm has solely highly parallel local operations and this property makes feasible parallel implementation suitable for very large scale integration (VLSI) design. Simulation on a number of video sequences is presented with comparison to BKM-ME and other fast block search algorithms for video coding applications even though DXT-ME is completely different from any block search algorithms.     

Layered motion estimation

A layered motion estimation algorithm is proposed that permits quasisimultaneous motion estimation/segmentation up to a fixed maximum number of layers. The estimation results in one motion parameter set per layer, and a segmentation map that assigns these sets to different parts of the image (motion layers). Motion in a layer is modelled with at maximum four parameters capable of describing pan, tilt and zoom. The concept shows some hierarchy, i.e. a ranking of the motion layers. In this way the motion parameter estimation concerning one layer excludes those parts of the image that have been described by a layer ranked higher in the hierarchy and are not polluted by parts of the image that are better described by layers ranked lower in the hierarchy. The concept results in a very low operations count. It has been shown to perform well even in critical scan rate conversion applications, particularly in picture rate up-conversion. A variant including three layers has been scheduled to run in real-time on a Philips TriMedia processor.     

Rate-optimised motion estimation

The authors propose a motion estimation (ME) algorithm for determining a motion vector which minimises the estimated total number of bits allocated to the motion vector and to the residual discrete cosine transform coefficients. This rate-optimised ME algorithm contributes to an improved reconstructed image quality and a reduced number of coding bits per frame     

Fast motion estimation for H.264

Several specific features have been incorporated into Motion estimation (ME) in H.264 coding standard to improve its coding efficiency. However, they result in very high computational load. In this paper, a fast ME algorithm is proposed to reduce the computational complexity. First, a mode discriminant method is used to free the encoder from checking the small block size modes in homogeneous regions. Second, a condensed hierarchical block matching method and a spatial neighbor searching scheme are employed to find the best full-pixel motion vector. Finally, direction-based selection rule is utilized to reduce the searching range in sub-pixel ME process. Experimental results on commonly used QCIF and CIF format test sequences have shown that the proposed algorithm achieves a reduction of 88% ME process time on average, while incurring only 0.033 dB loss in PSNR and 0.50% increment on the total bit rate compared with that of exhaustive ME process, which is a default approach adopted in the JM reference software.     

Fast motion estimation for surveillance video compression

In this article, novel approaches to perform efficient motion estimation specific to surveillance video compression are proposed. These includes (i) selective (ii) tracker-based and (iii) multi-frame-based motion estimation. In selective approach, motion vector search is performed for only those frames that contain some motion activity. In another approach, contrary to performing motion estimation on the encoder side, motion vectors are calculated using information of a surveillance video tracker. This approach is quicker but for some scenarios it degrades the visual perception of the video compared with selective approach. In an effort to speed up multi-frame motion estimation, we propose a fast multiple reference frames-based motion estimation technique for surveillance videos. Experimental evaluation shows that significant reduction in computational complexity can be achieved by applying the proposed strategies.     

Content-adaptive motion estimation algorithm for coarse-grain SVC

A joint model of scalable video coding (SVC) uses exhaustive mode and motion searches to select the best prediction mode and motion vector for each macroblock (MB) with high coding efficiency at the cost of computational complexity. If major characteristics of a coding MB such as the complexity of the prediction mode and the motion property can be identified and used in adjusting motion estimation (ME), one can design an algorithm that can adapt coding parameters to the video content. This way, unnecessary mode and motion searches can be avoided. In this paper, we propose a content-adaptive ME for SVC, including analyses of mode complexity and motion property to assist mode and motion searches. An experimental analysis is performed to study interlayer and spatial correlations in the coding information. Based on the correlations, the motion and mode characteristics of the current MB are identified and utilized to adjust each step of ME at the enhancement layer including mode decision, search-range selection, and prediction direction selection. Experimental results show that the proposed algorithm can significantly reduce the computational complexity of SVC while maintaining nearly the same rate distortion performance as the original encoder.