Time filter

Source Type

Yuan H.,Shandong University | Liu J.,Shandong University | Li Z.,Applied Technology Internet | Liu W.,Hisense State Key Laboratory of Digital Multi Media Technology Company
Electronics Letters | Year: 2012

Depth is one of the most important parameters in three dimensional videos (3DV). To improve the coding efficiency of depth maps, a virtual view oriented distortion criterion is proposed to substitute the mean squared error during the rate distortion optimisation process. By employing the proposed distortion criterion, the bit rates of depth maps could be reduced while maintaining the same quality of synthesised virtual views. Experimental results demonstrate that an average of -6.89 and a maximum of -23.04 coding bit rate reduction of depth maps could be achieved while maintaining the same quality of synthesised virtual view. © 2012 The Institution of Engineering and Technology.


Zhang B.,Shandong University | Liu J.,Shandong University | Ge J.,Shandong University | Yuan H.,Shandong University | Liu W.,Hisense State Key Laboratory of Digital Multi Media Technology Corporation
ICALIP 2012 - 2012 International Conference on Audio, Language and Image Processing, Proceedings | Year: 2012

This paper presents a video super resolution algorithm to enhance definition for a low resolution video sequence within a mixed spatial-temporal resolution stereo video. Two different views of the same scene with large overlapped area are captured by two cameras: one is low-resolution high-speed camera, whilst the other is high-resolution low-speed camera. First, the low resolution frames are spatially super-resolved by compensating new high-frequency information followed by stereo matching. Then the remainder frames of the low resolution video are super resolved via bidirectional overlapped block motion compensation using adjacent high resolution frames obtained above. Experimental results demonstrate that both the subjective and objective qualities of the super resolved videos are significantly improved by the proposed algorithm. © 2012 IEEE.


Nie X.-S.,Shandong University | Nie X.-S.,Shandong University of Finance and Economics | Liu J.,Shandong University | Liu J.,Hisense State Key Laboratory of Digital Multi Media Technology Corporation | And 2 more authors.
Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology | Year: 2011

A video hashing based on Locally Linear Embedding (LLE) is proposed in this paper. In this method, some representative frames are first selected based on a graph model, and four-order cumulants are taken as features of video in the high dimensional feature space. Then the video is mapped to a three-dimensional space using LLE, and video hash sequence is generated using the norms of points in the three-dimensional space to detect video copies. Experimental results show that the video hashing has good robustness and discrimination.


Wang D.,Shandong University | Liu J.,Shandong University | Sun J.,Shandong University | Liu W.,Hisense State Key Laboratory of Digital Multi Media Technology Co. | Li Y.,Hisense State Key Laboratory of Digital Multi Media Technology Co.
IEEE International Symposium on Broadband Multimedia Systems and Broadcasting, BMSB | Year: 2012

Semi-automatic 2D-to-3D conversion becomes very popular in 3D contents creation due to its advantages over balancing the tradeoff between labor cost and 3D conversion effect. However, the key-frame extraction, as a very important step, has not been specifically put forward in the existing systems. In this paper, a novel key-frame extraction method based on cumulative occlusion is proposed for 2D-to-3D system. An input long color video is first segmented into several shots by using the block-based histogram difference. Then shot filtering is performed by several principles which are specific in 2D-to-3D system. After shot segmentation, for each shot, the cumulative occlusion curve is computed, according to which key-frames are selected. Objective evaluation shows that, compared with the previous methods, the proposed key-frame selection algorithm can keep depth errors of all frames in the whole video controlled in a lower level. And the good propagated depth maps indicate that our proposed schemes can be used for most semi-automatic 2D-to-3D systems. © 2012 IEEE.


Zhang B.,Shandong University | Chen C.,Shandong University | Yan H.,Shandong University of Finance and Economics | Liu W.,Hisense State Key Laboratory of Digital Multi Media Technology Corporation
International Conference on Signal Processing Proceedings, ICSP | Year: 2010

Image registration is an important task in the field of computer vision and pattern recognition. In this paper, we propose a robust sub-pixel registration algorithm which is based on multi-resolution and new edge detection interpolation method. After applying truncated window function to the images to be registration, the low-pass bands of the wavelet decomposition are applied to build the image pyramid. Performance of our proposed algorithm is compared with other method using the same block matching algorithm and mutual information similarity measure. The results show that our proposed algorithm is effective for noisy degradation condition and can lead to highly accurate image registration. © 2010 IEEE.


Nie X.,Shandong University of Finance and Economics | Nie X.,Shandong University | Qiao J.,Shandong Normal University | Liu J.,Shandong University | And 4 more authors.
International Conference on Signal Processing Proceedings, ICSP | Year: 2010

As web video databases tend to contain immense copies with the explosive growth of online videos, effective and efficient copy identification techniques are required for content management and copyrights protection. To this end, this paper presents a novel video hashing for video copy identification based on Locally Linear Embedding (LLE). It maps the video to a lowdimensional space through LLE, which is invariant to translation, rotation and rescaling. In this way, we can use the points mapped from the video to play as a robust hashing. Meanwhile, to detect copies which are parts of original videos or contain a clip that comes from original. A dynamic sliding window is applied for matching. Experimental results show that the video hashing is of good robustness and discrimination. © 2010 IEEE.


Wang D.,Shandong University | Liu J.,Shandong University | Ren Y.,Shandong University | Ge C.,Shandong University | And 2 more authors.
2012 International Conference on Wireless Communications and Signal Processing, WCSP 2012 | Year: 2012

Semi-automatic 2D-to-3D conversion (2D-3D) is preferred due to its advantage of handling the trade-off between human participation and 3D conversion effects. In this paper, a novel depth propagation algorithm based on depth consistency is proposed, which can be widely employed in semi-automatic 2D-3D. The depth consistency refers to the principle that two neighboring pixels should have similar depth values if their color values or intensities are similar. Based on this observation, depth estimation is modeled as a constrained optimization problem. Two contributions of this paper are : first, a convenient tool for image 2D-3D is implemented, which can obtain a good depth map with the help of limited user's scribbles on the input image; second, a novel depth propagation algorithm is presented for estimating the depth maps of the non-key-frames in video 2D-3D. In this algorithm, the depth values of high-confidence matched pixels in the non-key-frame are first assigned with the values of the correspondences in the key-frame and then, depth values of the remaining pixels are estimated by solving a constrained optimization problem which is constructed by exploiting depth consistency. Experimental results show that, compared with the shifted bilateral filtering (SBF) algorithm, the proposed algorithm not only holds a similar performance, but also increases the processing speed by nearly 5 times on average. © 2012 IEEE.


Wu Q.,Shandong University | Liu J.,Shandong University | Liu J.,Hisense State Key Laboratory of Digital Multi Media Technology Co. | Sun J.,Shandong University | And 2 more authors.
Journal of Computational Information Systems | Year: 2012

In this paper, we propose a new shift-invariant feature extraction method based on tensor analysis for robust text-independent speaker Identification. Multiple factors including time, frequency, scale and phase are investigated to extract the essential features of speech signal. In order to explore the shift-invariant characteristic of spectral features, we propose a convolutive model to learn the spectral basis function in tensor structure. Nonnegative assumption is also imposed on the CANDECOMP model to ensure the sparsity and preserve robust features. Experimental results demonstrate that our proposed method can improve the recognition accuracy specifically in noise conditions. © 2012 Binary Information Press.


Nie X.,Shandong University | Nie X.,Shandong University of Finance and Economics | Liu J.,Shandong University | Sun J.,Shandong University | Liu W.,Hisense State Key Laboratory of Digital Multi Media Technology Corporation
IEEE Signal Processing Letters | Year: 2011

A robust video hashing scheme for video content identification and authentication is proposed, which is called Double-Layer Embedding scheme. Intra-cluster Locally Linear Embedding (LLE) and inter-cluster Multi-Dimensional Scaling (MDS) are used in the scheme. Some dispersive frames of the video are first selected through graph model, and the video is partitioned into clusters based on the dispersive frames and the K-Nearest Neighbor method during the hashing. Then, the intra-cluster LLE and inter-cluster MDS are used to generate local and global hash sequences which can inherently describe the corresponding video. Experimental results show that the video hashing is resistant to geometric attacks of frames and channel impairments of transmission. © 2006 IEEE.


Patent
Hisense State Key Laboratory Of Digital Multi Media Technology Co. | Date: 2012-03-07

A motion detection method, apparatus and system are disclosed in the present invention, which relates to the video image processing field. The present invention can effectively overcome the influence of the background on motion detection and the problem of object conglutination to avoid false detection, thereby accomplishing object detection in complex scenes with a high precision. The motion detection method disclosed in embodiments of the present invention comprises: acquiring detection information of the background scene and detection information of the current scene, wherein the current scene is a scene comprising an object(s) to be detected and the same background scene; and calculating the object(s) to be detected according to the detection information of the background scene and the detection information of the current scene. The present invention is applicable to any scenes where moving objects need to be detected, e.g., automatic passenger flow statistical systems in railway, metro and bus sectors, and is particularly applicable to detection and calibration of objects in places where brightness varies greatly.

Loading Hisense State Key Laboratory of Digital Multi Media Technology Corporation collaborators
Loading Hisense State Key Laboratory of Digital Multi Media Technology Corporation collaborators