block matching algorithm (BMA) is the core of Motion-Compensated Frame Interpolation (MCFI), and its accuracy greatly affects the interpolation quality of MCFI. To improve BMA accuracy, this paper proposes the use of ...
详细信息
block matching algorithm (BMA) is the core of Motion-Compensated Frame Interpolation (MCFI), and its accuracy greatly affects the interpolation quality of MCFI. To improve BMA accuracy, this paper proposes the use of a self-similarity based context feature to improve the matching accuracy of BMA. First, we extract the patch centered at any pixel in a block, and perform the self-similarity descriptor to generate its correlation surface. Second, the correlation surface is statistically measured to represent the context feature, and the context cube of a block is produced by attaching the corresponding context feature to each pixel. Finally, we fuse the context cube into bidirectional matching criterion of BMA to get the motion vector field of the absent frame, and predict the absent frame by using motion compensation interpolation. Experimental results show that the proposed algorithm improves the BMA accuracy with a low computational complexity, and is better than the traditional MCFI algorithms in terms of both objective and subjective quality of the interpolated frames.
Motion Estimation in a video is a challenging optimization problem where the objective is to minimize the error or to maximize the correlation between a Macro block (MB) in current frame and MB in the reference frame ...
详细信息
Motion Estimation in a video is a challenging optimization problem where the objective is to minimize the error or to maximize the correlation between a Macro block (MB) in current frame and MB in the reference frame and to find the best matching MB. As the pace of development of conventional Pattern-Based block matching algorithms was very fast and also so many in numbers that it almost got saturated. But at the same time rapid development of new algorithms based on natural genetics to exploit intelligence such as genetic algorithms, evolutionary algorithm, particle swarm optimization and differential evolution show the potential of these algorithms in optimization and gave researchers a broad dimension to apply such nature inspired algorithms in the field of motion estimation. Subsequently, researchers started implementing one by one many nature-inspired algorithms, sometimes referred to as soft computing techniques, for solving the optimization problem of motion estimation and proved that soft computing techniques have huge advantages over the conventional pattern-based or prediction based block matching algorithms. In this paper, several nature-inspired algorithms that are implemented for video motion estimation are reviewed and the performance is compared to highlight the competitive advantages of soft computing techniques over existing fixed pattern search algorithms. (C) 2019 Karabuk University. Publishing services by Elsevier B.V.
This work presents the implementation of a matching-based motion estimation sensor on a Field Programmable Gate Array (FPGA) and NIOS II microprocessor applying a C to Hardware (C2H) acceleration paradigm. The design,...
详细信息
This work presents the implementation of a matching-based motion estimation sensor on a Field Programmable Gate Array (FPGA) and NIOS II microprocessor applying a C to Hardware (C2H) acceleration paradigm. The design, which involves several matchingalgorithms, is mapped using Very Large Scale Integration (VLSI) technology. These algorithms, as well as the hardware implementation, are presented here together with an extensive analysis of the resources needed and the throughput obtained. The developed low-cost system is practical for real-time throughput and reduced power consumption and is useful in robotic applications, such as tracking, navigation using an unmanned vehicle, or as part of a more complex system.
In this paper, we propose a multi-level block matching algorithm using motion information in blocks. In the proposed algorithm, the block-level is decided by the motion degree in the block before motion searching proc...
详细信息
In this paper, we propose a multi-level block matching algorithm using motion information in blocks. In the proposed algorithm, the block-level is decided by the motion degree in the block before motion searching procedure, and then adequate motion searching performs according to the block-level. Which improves computational efficiency by eliminating an unnecessary searching process in no motion or low motion regions, and brings more accurate estimation results by deepening motion searching process in high motion regions. Simulation results show that the proposed algorithm brings the lower estimation error about 20% MSE reduction - with the fewer blocks per frame and the lower computational loading - about 98% operational amount reduction - than full search block matching algorithm with constant block size.
This paper presents a new motion segmentation method using iterative maximum likelihood framework. The method consists of two steps. In the first step, motion regions are detected and motion vectors are computed for t...
详细信息
This paper presents a new motion segmentation method using iterative maximum likelihood framework. The method consists of two steps. In the first step, motion regions are detected and motion vectors are computed for these detected regions. In the second step, a similarity matrix is Computed from the motion vectors and motion segmentation is done using maximum likelihood method. The method has been tested using real world motion sequences and is found to give very low error rate and it also detects the slow moving objects.
A merging procedure joining search pattern and variable block size motion estimation for H.264/AVC is proposed in this paper. The principal purpose of the proposed methods is the reduction of the computational complex...
详细信息
A merging procedure joining search pattern and variable block size motion estimation for H.264/AVC is proposed in this paper. The principal purpose of the proposed methods is the reduction of the computational complexity for blockmatching module. In fact, there are numerous contributions in the literature aiming the reduction of the computational cost needed for motion estimation. The best solution from a qualitative point of view is the full search that considers every possible detail. The computational effort required is enormous and this makes motion estimation by far the most important computational bottleneck in video coding systems. Our approach invests and exploits the center-biased characteristics of the real world video sequences, aiming to achieve an acceptable image quality while independently targeting the reduction of the computational complexity. The simulations results demonstrated that the proposal performs well.
In this venture, an imaginative unpredictability movement that has low vector preparing calculation toward the end side is proposed for movement repaid video vector frame interpolation. By handling this calculation, w...
详细信息
In this venture, an imaginative unpredictability movement that has low vector preparing calculation toward the end side is proposed for movement repaid video vector frame interpolation. By handling this calculation, we typically demonstrates the issues of broken edges and disfigured structures in an inserted frame by progressively refining movement vectors on various square sizes. In the proposed strategy, the info has been taken as video rather than pictures in the current framework and the recuperation yield is taken as pictures and further process has been experienced to get the yield as video. There are some extraordinary systems in this strategy, for example, stage-based interpolation method, multistage movement repaid interpolation, and so forth is usually used to get high reason picture with diminished haze in the pictures to get the unmistakable picture of the information videos. Exploratory outcomes will demonstrate the visual quality, toughness in the video groupings containing quick moving video.
To reduce an amount of computation of full search algorithm for fast motion estimation, we propose a new and fast matchingalgorithm without any degradation of predicted images. The computational reduction without any...
详细信息
To reduce an amount of computation of full search algorithm for fast motion estimation, we propose a new and fast matchingalgorithm without any degradation of predicted images. The computational reduction without any degradation comes from adaptive matching scan algorithm according to the image complexity of the reference block in current frame. Experimentally, ae significantly reduce the computational load compared with conventional full search algorithm.
In this paper, we propose a motion vector correction algorithm to correct the erroneous motion vectors due to the failure of conventional motion estimation (ME) based on block matching algorithm (BM,4). The proposed m...
详细信息
In this paper, we propose a motion vector correction algorithm to correct the erroneous motion vectors due to the failure of conventional motion estimation (ME) based on block matching algorithm (BM,4). The proposed method is based on the pattern-like image analysis that consists of the periodic pattern detection and calculation of motion parameters for the motion vector correction. In addition, the proposed method corrects effectively the erroneous motion vectors using the relations between the periodic pattern and local/global minima depending on whether the periodic pattern is static or moving. Also, it does not significantly increase computational complexity since it uses the estimated motion vectors and the integral projection of the sum of absolute difference (SAD) previously obtained from the conventional ME. Experimental results indicate good performance of the proposed scheme with significant reduction the erroneous motion vectors.
作者：
Kim, JMAhn, BHKBS
Tech Res Inst Yongdeungpo Gu Seoul 150790 South Korea K JIST
Dept Mechatron Puk Gu Kwangju 500712 South Korea
We propose a new and fast full search (FS) motion estimation algorithm for video coding. The computational reduction comes from sequential rejection of impossible candidates with derived formula and subblock norms. Ou...
详细信息
We propose a new and fast full search (FS) motion estimation algorithm for video coding. The computational reduction comes from sequential rejection of impossible candidates with derived formula and subblock norms. Our algorithm reduces more the computations than the recent fast full search (FS) motion estimation algorithms.
暂无评论