Fast Half Pixel Motion Estimation Based on Spatio-temporal Correlations

A prediction-based simplified half pixel motion estimation algorithm is proposed to improve the search speed in half pixel motion estimation. It predicts the possible sub area and checks only three half pixel points in its possible sub area rather than eight half pixel points which are used in full half pixel search method (FHPS). Experimental results show that the proposed algorithm has a significant speedup up to 62% over FHPS and preserves the image quality

H.264/AVC is the outstanding and significant video compression standard developed by ITU-T/ISO/IEC Joint Video Team (JVT). Motion estimation (ME) plays a key role in H.264/AVC, it concerns greatly on computational complexity especially when using the full search (FS) algorithm. Although many fast ME algorithms have been proposed to reduce the huge calculation complexity instead of FS, the ME still can not satisfy the critical real-time applicationpsilas needs. In this paper, a fast integer pixel variable block motion estimation algorithm based on JVT which accepted UMHexagonS algorithm is presented for H.264/AVC encoder. With special considerations on the motion activity of the current macro-block, several techniques, i.e., adaptive search strategies have been utilized to significantly improve the video coding performance. The simulation results analysis shows that the proposed algorithm maintains an unnoticeable quality loss in terms of PSNR on average compared with FS and reduces nearly 20% motion estimation time compared with UMHexagonS while maintaining coding efficiency.

To study the effect of decentration and tilt of the type I Boston keratoprosthesis (KPro) on image quality in both aphakic and pseudophakic eyes. An optical ray-tracing program was used to simulate the image projected onto the retina in an eye with a perfectly centered KPro, and in eyes with varying degrees of KPro decentration and tilt. Decentration was modeled along a typical white-to-white distance of 12.0 mm, and the corresponding tilt was calculated assuming a radius of curvature of 8.0 mm, the radius of curvature of the backplate of the KPro. Both aphakic and pseudophakic eyes were simulated, and the corresponding modulation transfer function curves, point spread functions, and astigmatism were analyzed. The perfectly centered KPro produced a high-quality image with no induced astigmatism. Increasing decentration beyond approximately 0.5 mm resulted in poorer image quality with a more pronounced effect in the presence of an intraocular lens. Using models of the normal eye as a threshold, image degradation due to decentration becomes clinically significant at approximately 1.4 mm and 0.9 mm for the aphakic and pseudophakic cases, respectively. Astigmatism introduced by decentration is approximately 0.25 D cylinders at those thresholds. Decentration of up to 0.5 mm had no significant impact on image quality and an attempt at good intraoperative centration of the KPro within this range is important. Conversely, decentration of 0.9 mm or more during surgical implantation can result in significant degradation in retinal image quality including astigmatism. The effect is greater in the pseudophakic eye.

Most of the fast search motion estimation algorithms reduce the computational complexity of motion estimation (ME) greatly by checking only a few search points inside the search area. In this paper, we propose a new algorithm--multi-layer motion estimation (MME) which reduces the computational complexity of each distortion measure instead of reducing the number of search points. The conventional fast search motion estimation algorithms perform ME on the reference frame with full distortion measure; on the contrary, the MME performs ME on the layers with partial distortion measures to enhance the computational speed of ME. A layer is an image which is derived from the reference frame; each macro-pixel value in the layer represents the sum of the values of the corresponding pixels in the reference frame. A hierarchical quad-tree structure is employed in this paper to construct multiple layers from the reference frame. Experimental results on different video sequences show evidence that many motion vectors have been found similar both in the reference frame and the layers. The effectiveness of the proposed MME algorithm is compared with that of some state-of-the-art fast block matching algorithms with respect to speed and motion prediction quality. Experimental results on a wide variety of video sequences show that the proposed algorithm outperforms the other popular conventional fast search motion estimation algorithms computationally while maintaining the motion prediction quality very close to the full-search algorithm. Moreover, the proposed algorithm can achieve a maximum of 97.99 % speed-improvement rate against the fast full-search motion estimation algorithms which are based on hierarchical block matching process. The proposed MME performs the motion estimation on the layers by using three types of search patterns. The derivation of these search patterns exploits the characteristic of the center-biased motion vector distribution and that of less intensive block distortion measurement of the layers.

In an H.264 video encoder, motion estimation (ME) is the most time-consuming component. The ME process consists of two stages, integer pixel search and fractional pixel search. Since the complexity of integer pixel search has been greatly reduced by numerous fast ME algorithms, the computation overhead required by fractional pixel ME has become relatively significant. To reduce the complexity of fractional pixel ME, we propose a prediction-based directional fractional pixel ME algorithm. We utilize more accurate motion vector predictions and directional search to achieve better computation reduction. We further propose an early termination method to decrease the amount of search. Experimental results show that, compared to the full search sub-pel ME and the fast sub-pel ME proposed in H.264, the proposed method can reduce up to 84% and 74% of fractional pixel search points respectively, with a negligible degradation in quality.

In this paper, we propose a variable step search fast motion estimation algorithm using local statistics of neighboring motion vectors. Using the degree of correlation between neighboring motion vectors, motion search range of each macroblock is adaptively adjusted to reduce unnecessary search points. Based on the determined search range, motion vector is obtained by variable step search step. Experimental results show that comparing to fast full spiral search, fixed step, and fast variable block size motion estimation algorithms, the motion search points and computing time of proposed algorithm are dramatically reduced with acceptable loss in coding performance.

To evaluate metal artefact reduction (MAR) techniques from four CT vendors in hip prosthesis imaging. Bilateral hip prosthesis phantom images, obtained by using MAR algorithms for single-energy CT data or dual-energy CT (DECT) data and by monoenergetic reconstructions of DECT data, were visually graded by five radiologists using 10 image quality criteria. Comparisons between the MAR images and a reference image were performed for each scanner separately. Ordinal probit regression analysis was used. The MAR algorithms in general improved the image quality based on the majority of the criteria (up to between 8/10 and 10/10) with a statistical improvement in overall image quality (p < 0.001). However, degradation of image quality, such as new artefacts, was seen in some cases. A few monoenergetic reconstruction series improved the image quality (p < 0.004) for one of the DECT scanners, but it was only improved for some of the criteria (up to 5/10). Monoenergetic reconstructions resulted in worse image quality for the majority of the criteria (up to 7/10) for the other DECT scanner. The MAR algorithms improved the image quality of the hip prosthesis CT images. However, since additional artefacts and degradation of image quality were seen in some cases, all algorithms should be carefully evaluated for every clinical situation. Monoenergetic reconstructions were in general concluded to be insufficient for reducing metal artefacts. Qualitative evaluation of the usefulness of several MAR techniques from different vendors in CT imaging of hip prosthesis.

Most of the fast search motion estimation algorithms reduce the computational cost of motion estimation greatly by checking only a few search points inside the search area by using full distortion measure. This paper proposes a multi-layer motion estimation which employs partial distortion as its distortion measure to reduce the number of computations involved in each search point instead of reducing the number of search points. A layer is an image which is derived from the reference frame such that the sum of a block of pixels in the reference frame determines the point of a layer. It has been noticed on different video sequences that many motion vectors on the layers are the same as those searched on the reference frame. Experimental results on a wide variety of video sequences show that the proposed algorithm outperforms the other popular conventional fast search motion estimation algorithms computationally while maintaining the motion prediction quality very close to the full-search algorithm.

In single viewpoint video coding, there are many fast block matching motion estimation algorithms proposed, such as three-step search, four-step search, diamond search as well as hexagon-based search and so on. However, experimental analysis show that these algorithms are not suitable for using directly in Multiview View Coding (MVC). Since the increased search range, the larger format of multiview video as well as the correlations between inter-view frames are not considered by these algorithms, they may easy led the block matching search into local minimum, the Rate-Distortion (R-D) performance will degrade dramatically. In this paper, we propose a novel multiple hexagon search algorithm to address this problem. Firstly, according to the original initial search point, four sub-search windows are constructed. Then, the hexagon based search algorithm will be performed respectively in the four sub-search windows. The final result is the best search point with the minimum R-D cost among the best points in the four sub-search windows. In order to trade off the computational complexity and R-D performance, two adaptive early termination strategies are proposed. The experimental results show that the proposed algorithm yields a quite promising coding performance in terms of R-D performance and computational complexity. Especially, the proposed algorithm can work well in multiview video sequences with various motion and disparity activities.

Predictive fast Motion Estimation (ME) algorithms have been widely used in video CODECs due to their performance efficiency and low computational complexity. In this thesis, a new block-based fast motion estimation technique named Dynamic Predictive Search Algorithm (DPSA) is developed, which can be considered in predictive zonal search category. The proposed approach is based on the observation that temporally and spatially adjacent macro-blocks are not just statically correlated, but also dynamic alterations in their motion content are highly coherent. DPSA introduces a new set of six candidate predicted motion vectors. For early termination criteria, DPSA modifies termination procedure of already existing EPZS algorithm. Performance of this newly proposed algorithm has been compared to four other state-of-the-art algorithms implemented on JVT, H.264 standard software platform. Experimental results have proven that DPSA accomplishes up to 38% compression ratio enhancement achieved by a process with more 14.75% less computational complexity and up to0.47 dB higher PSNR values over the EPZS. It also manages to have up to 13% speed up over EPZS algorithm. Because of its simplicity and low computational complexity DPSA is energy efficient for portable video processing in computation- or power-constrained applications and easy to be implemented on both FPGA- and Microcontroller-based embedded systems. Also, higher compression ratio makes DPSA more compatible with limited capacity storage media, and limited band-width transmission networks.

Background: Guidelines for SPECT myocardial perfusion imaging (MPI) traditionally recommend a fixed tracer dose. Yet, clinical practice shows degraded image quality in heavier patients. The aim was to optimize and validate the tracer dose and scan time to obtain a constant image quality less dependent on patients' physical characteristics. Methods: 125 patients underwent Cadmium Zinc Telluride (CZT)-SPECT stress MPI using a fixed Tc-99m-tetrofosmin tracer dose. Image quality was scored by three physicians on a 4-point grading scale and related to the number of photon counts normalized to tracer dose and scan time. Counts were correlated with various patient-specific parameters dealing with patient size and weight to find the best predicting parameter. From these data, a formula to provide constant image quality was derived, and subsequently tested in 92 new patients. Results: Degradation in image quality and photon counts was observed for heavier patients for all patients' specific parameters (p<0.01). We found body weight to be the best-predicting parameter for image quality and derived a new dose formula. After applying this new body weight-depended tracer dose and scan time in a new group, image quality was found to be constant (p>0.19) in all patients. Conclusions: Also in CZT SPECT image quality decreases with weight. The use of a tracer dose and scan time that depends linearly on patient's body weight corrected for the varying image quality in CZT-SPECT MPI. This leads to better radiation exposure justification.

For video compression, motion estimation is popularly employed to exploit temporal correlation existing in video sequences. If we employ the full search block matching algorithm for estimating motion vectors, it requires very heavy computational complexity. Although several fast block matching algorithms have been proposed to solve this problem, they sacrifice their reconstructed image quality. In this paper, we derive optimal search patterns for fast block matching motion estimation. By analyzing the block matching algorithm as a function of the block size and the shape, we find optimal search patterns for initial motion estimation. The proposed idea can provide an analytical ground for the current MPEG-2 proposals. In addition, we propose a new fast motion estimation algorithm using adaptive search patterns, considering matching criteria and statistical properties of object displacement. In order to select an appropriate search pattern, we exploit the relationship between the motion vector and the frame difference of each block. By changing the search pattern adaptively, we can improve the motion prediction accuracy, while reducing the required computational complexity compared to other fast block matching algorithms.

H.264 is a new recommendation for moving picture coding proposed by JVT. In this standard, full pixel and sub-pixel motion estimation are the most important parts. The unequal-arm adaptive rood pattern search (APRS-3) algorithm has shown a good performance on search speed-up. In this paper, we propose an improved algorithm called APRS-4 which applies an early termination technology on APRS-3 by using adaptive threshold. This early termination technology is also applied on sub-pixel motion estimation, and a new fast sub-pixel motion estimation algorithm is proposed which uses the motion vector information of the neighbor blocks. Experimental results show that both algorithms have achieved a good speed-up ratio. Compared with full-pixel motion estimation, our new algorithm could keep image quality well with almost the same bit rate.

There is an increasing need for high quality video on low power, portable devices. Possible target applications range from entertainment and personal communications to security and health care. While H.264/AVC answers the need for high quality video at lower bit rates, it is significantly more complex than previous coding standards and thus results in greater power consumption in practical implementations. In particular, motion estimation (ME), in H.264/AVC consumes the largest power in an H.264/AVC encoder. It is therefore critical to speed-up integer ME in H.264/AVC via fast motion estimation (FME) algorithms and hardware acceleration. In this paper, we present our hardware oriented modifications to a hybrid FME algorithm, our architecture based on the modified algorithm, and our implementation and prototype on a PowerPC-based Field Programmable System on Chip (FPSoC). Our results show that the modified hybrid FME algorithm on average, outperforms previous state-of-the-art FME algorithms, while its losses when compared with FSME, in terms of PSNR performance and computation time, are insignificant. We show that although our implementation platform is FPGA-based, our implementation results compare favourably with previous architectures implemented on ASICs. Finally we also show an improvement over some existing architectures implemented on FPGAs.

Motion estimation (ME) for the high efficiency video coding (HEVC) occupies the largest part in the computational complexity of the HM encoder, which consists of integer ME (iME) and fractional ME (fME). We propose a fast fME algorithm that reduces its complexity substantially by reducing the number of search points to less than a half, based on statistical observations, in which four search points are selected only in the horizontal and vertical directions for half-pel search and three search points that have the smaller absolute value of the motion vector difference are selected from eight candidates for quarter-pel search. The effective number of search points in the proposed fME algorithm is reduced to about 60% and the interpolation filtering is also reduced to about 55%, compared to that of the fME in the HM encoder while experimental results show that its Bj⊘ntegaard delta (BD) rate degradation is 0.93% for the five class-B video sequences.