Dong-Keun Lim

A fast block matching motion estimation algorithm based on statistical properties of object displacement

For video compression, motion estimation is popularly employed to exploit the temporal correlation existing in video sequences. When a full search (FS) block matching algorithm (BMA) is used for estimating motion vectors, it requires very heavy computational complexity. Although several fast block matching algorithms have been proposed to solve this problem, those methods sacrifice their reconstructed image qualities. We propose a new motion estimation algorithm by considering matching criteria and statistical properties of object displacement. We exploit the relationship between the motion and frame difference of each block to choose a more compact search pattern for BMA. By changing the search pattern adaptively, we can improve motion prediction, while reducing computational complexity than other fast BMA algorithms.

Multiresolution motion compensation in the wavelet domain for scalable video coding

Wavelet transform is a popular tool for image and video coding. It has several advantages in multiresolution analysis and subband decomposition of images. Motion estimation and motion compensation methods are widely used to reduce temporal redundancy in video sequences. Although there have been several attempts to compensate for the motion in the wavelet domain, their performances are limited due to the shift-variant problem. In this paper, we propose a new motion compensation method in the wavelet domain to overcome the shift-variant problem. Experimental results show that the proposed method outperforms the previous motion compensation methods in the wavelet domain as well as the full search algorithm in the spatial domain.

Image segmentation using hierarchical meshes

The object boundary of an image plays an important role in image interpretation. In this paper, we introduce a concept of hierarchical mesh-based image segmentation for finding object boundaries. In each hierarchical layer, we employ neighborhood searching and boundary tracking methods to refine the initial boundary estimate. We also apply a local region growing method to define closed contours. Experimental results indicate that reliable segmentation of objects can be accomplished by the proposed low complexity technique.

Adaptive Motion Search Based on Block Difference and Motion Magnitude

In this paper, we derive optimal search patterns for fast block matching motion estimation. Since the motion search pattern is important in terms of search speed and correctness of the motion information, we consider various search patterns and search strategies. By analyzing the block matching algorithm as a function of the block size and distance in the search area, we find analytic search patterns for initial motion estimation. We also propose an adaptive motion search algorithm, where we exploit the correlation between block difference and motion magnitude. The proposed idea can provide an analytical ground for the MPEG-4 algorithms for fast motion search. We can improve the prediction accuracy of motion estimation, while reducing the required computational complexity compared to other fast block matching algorithms.

Image warping using adaptive partial matching

In this paper, we propose an adaptive partial matching method for motion estimation to reduce the computational complexity, while maintaining the image quality comparable to the hexagonal matching method. The proposed motion compensation method combines the fast affine transformation using a vector relationship. We simulate our proposed motion estimation method in a DCT-based coder by encoding CIF (common intermediate format) images at the bitrate of below 64 kb/s. The quality of reconstructed image with our method is substantially improved compared with the block matching algorithms (BMA), and is comparable to the hexagonal matching method. Computational complexity and coding bits are also reduced significantly relative to the BMA and the conventional image warping methods.

Segmentation of interest objects using the hierarchical mesh structure

The object boundary of an image plays an important role for image analysis and interpretation. The watershed algorithm and the region growing algorithm are popularly employed for image segmentation. These give reasonable performances, but require a large amount of computation time and sometimes fail to obtain continuous linkage of object boundary. In this paper, we introduce hierarchical mesh-based image segmentation. In each hierarchy, we employ neighborhood searching and boundary tracking methods to refine the initial boundary estimate. The proposed algorithm increases the robustness of linkage of object boundaries by overlooking and estimating connectivity and gives new modified chain coding. Reliable segmentation of objects can be accomplished by the proposed low complexity technique.

Fast image warping using adaptive partial matching

The block-matching motion estimation algorithm using a translational motion model cannot provide acceptable image quality in low bit-rate coding. To improve coding performance, we can use image warping using affine transformation as a more complicated motion model than the translational motion model. However, when some node points are flipped over in the image warping method, it creates severe deformations of the mesh structure and brings image degradations. It also requires large computational complexity. We show that wide search ranges do not always improve reconstructed image quality due to the large deformation of meshes. We analyze optimal search ranges according to frame difference and decide variable search ranges adaptively to get a motion vector at each node point. Since the block with a larger error than a threshold makes a large distortion, we give higher priority during motion estimation at a certain rate. To reduce computational complexity, we also introduce an adaptive partial matching method instead of applying the hexagonal matching method on the whole image. As a result, we develop an effective image warping method in terms of computational complexity, reasonable reconstructed image quality, and fewer number of coding bits.

A Fast Block-Matching Motion Estimation Algorithm with Motion Modeling and Motion Analysis

By modeling the block-matching algorithm as a function of the correlation of image blocks, we derive search patterns for fast block-matching motion estimation. The proposed approach provides an analytical support for the diamond-shape search pattern, which is widely used in fast block-matching algorithms. We also propose a new fast motion estimation algorithm using adaptive search patterns and statistical properties of the object displacement. In order to select an appropriate search pattern, we exploit the relationship between the motion vector and the block differences. By changing the search pattern adaptively, we improve motion prediction accuracy while reducing required computational complexity compared to other fast block-matching algorithms.

Fast-block-matching motion estimation algorithm using optimal search patterns

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.

Adaptive Partial Matching using Fast Spatial Transformation

In this paper, we examine new motion compensation methods based on the affine or bilinear transformation and derive fast algorithms for affine and bilinear transformation using vector relationship. We also develop a more effective motion estimation method than the conventional image warping method in terms of computational complexity, reconstructed image quality, and the number of coding bits. The performance of the proposed motion compensation method, which combines the affine or the bilinear transformation with the proposed adaptive partial matching, is evaluated experimentally. We simulate our proposed motion compensation method in a DCT- based coder by encoding CIF (Common Intermediate Format) images at bitrates of below 64 kb/s. The proposed adaptive partial matching method can reduce the computational complexity below about 50% of the hexagonal matching method, while maintaining the image quality comparable to the hexagonal method.