Long Ye

No-reference objective stereo video quality assessment based on visual attention and edge difference

No-reference objective stereo video quality assessment plays an important role in stereo video content production, especially to the sense of 2D-to-3D video conversion. The main problem needed to consider when designing an objective quality assessment model is to make the assessment result close to human perception. In this paper, a no-reference objective stereo video quality assessment method is proposed by taking consideration of visual attention and edge difference. For every frame in the given stereo video, the human interested regions and edge information are extracted separately. The edge information is used to calculate the blockiness, zero-crossing and disparity features for every block in the frames, and visual interested information determined the weights of each image block in the final assessment result calculation. Experimental results demonstrated the effectiveness of our proposed objective stereo video quality assessment method.

An improved SIFT algorithm based on adaptive threshold canny

The traditional SIFT algorithm is popular to extract the feature points of target objects, but it also brings the feature points of non-target objects together, leading to mismatching. This paper proposes an improved SIFT algorithm based on adaptive threshold canny operator. In the proposed method, since it has the advantages of accurate edge detection and anti-noise ability, the adaptive threshold canny operator is first employed to detect the edges of an image, and then we can find the feature points by SIFT in the peripheral region of the edges. By introducing the adaptive threshold canny operator, the target objects can be separated from background, largely increasing the matching rate of SIFT algorithm. Experimental results demonstrate that, compared with the traditional SIFT and SURF algorithms, the proposed method can improve the robustness of feature points and further increase the matching rate, meanwhile reducing the cost of running time in a certain extent.

Design of oversampled nonuniform filter banks with arbitrary rational frequency partitioning

Among the theory and design of oversampled nonuniform filter banks (NUFBs), most of the existing works only consider the cases with integer decimators, and up to now the issue with rational sampling factors has not been discussed yet. In this paper, we generalize the partial modulation technique to realize arbitrary rational frequency partitioning of oversampled NUFBs with highly desired linear-phase (LP) property. Further for the subbands with sampling factors violating the guard band restriction, a phase-modification scheme is derived to avoid uneliminable large aliasing and meanwhile preserving the LP characteristics of shifted analysis/synthesis filters. By using the proposed method, the design issue of LP oversampled NUFBs can be reduced to that of several prototypes, decreasing the design complexity largely. As illustrated by examples, the proposed algorithm is more general in terms of arbitrary rational decimation and thus has broad application prospects.

A Modified Joint Trilateral Filter for Depth Image Super Resolution

Depth image can be easily obtained by RGB-D sensors such as Microsoft Kinect, but the low resolution and poor quality of the obtained results pose a notable challenge on practical applications. To solve this problem, this paper proposes an algorithm of modified joint trilateral filter for depth image super resolution. In the proposed method, considering less texture contained in depth image, the high resolution (HR) edge is first extracted from its corresponding HR color image and then introduced to guide the modified joint trilateral filter primarily. Meanwhile, the intensity information is taken into account to avoid the fake edges in the HR edge map. With the guidance of HR edge and intensity information, the HR depth image could be simply interpolated via the modified joint trilateral filter. The experimental results manifest that our approach could not only save the running time but also obtain better performance compared with the state-of-the-art methods.

Camera Calibration Based Multi-Objects Location in Binocular Stereo Vision

In order to capture high quality binocular stereo video, it is necessary to manipulate both the convergence and the interaxial to take control of the depth of objects within the 3D space. Therefore the scene understanding becomes important as it can increase the efficiency of parameters control. In this paper, a camera calibration based multi-objects location method is introduced with the motivation that supply prior information of adjusting the convergence and the interaxial during capturing. Firstly, we are intended to calibrate the two cameras to get the intrinsic and extrinsic parameters. And then, we select points of the object in the images taken by left and right cameras respectively to determine its locations in the two images. With three-dimensional coordinate of objects, the distance between the object and camera baseline is calculated by mathematical methods.

A Novel Particle Filtering Framework Using Genetic Monte Carlo Sampling

Particle degeneration is a key issue in the performance of a particle filter. In this paper we introduce genetic Monte Carlo into sampling process with the basic idea of solving particle degeneration by means of evolution thought. It is shown that the novel particle filtering framework can effectively eliminate particle degeneration and reduce its dependency on the particle validity. Furthermore, the new genetic particle filter can be optimized by three key genetic factors-selection, crossover and mutation probabilities. select individual, which is analogical with genetic operations (select, crossover and mutation). The essence of genetic operations shows that it is possible to use genetic operators to sample in the data space; this is also our motivation to bring out genetic Monte Carlo sampling. This paper introduces genetic Monte Carlo sampling method, and then uses it in the resampling step of particle filter with the basic idea of solving particle degeneration problem by means of evolution thought. Furthermore, in order to make the systematic performance to reach the maximum of accuracy and enhance the tracking robustness, we propose an approach for searching the best solution by adjusting the probabilities of select, crossover and mutation operation. Finally, we illustrate the operation of our framework by an example: One- dimensional growth model. Result demonstrates the effectiveness of our method comparing with the other particle filtering frameworks.

Track Articulated Body Motion by Pose-Model Based Stochastic Search

Articulated body motion tracking remains a challenge problem as its high-dimension and the dependence between the joint freedoms. In this paper we present a novel framework to achieve articulated body motion (walking in this paper) tracking by the use of pose-model based stochastic search algorithm which partitions the state space into two pars(pose and noise) and then processes recursively. Besides, we also use marker-based Initialization to balance the limited information of the video, use path reform to improve the overall tracking performance. Experiment demonstrates the effectiveness of our method in solving the 2D articulated human movement tracking problem.

Design of linear-phase nonsubsampled nonuniform directional filter bank with arbitrary directional partitioning

Abstract In this paper, we propose a design method for linear phase (LP) nonsubsampled nonuniform directional filter bank (NUDFB) with arbitrary number of subbands and arbitrary directional partitioning. The proposed NUDFB is simply designed by windowing the analytical expressions of wedge-shaped filters in space domain. The direction and angular bandwidth of the filters are determined by only two angular parameters. It can extract directional information according to the directional distribution of images, making it efficient in the directional representation of images. In addition, the perfect reconstruction conditions are derived. Numerical experiments on image directional information extraction and image denoising are given to illustrate the performance of our NUDFB. The results show that our NUDFB outperforms various directional decomposition methods while possesses LP property and more flexibility.

A Real-Time Interactive System Based on Hand Gesture Recognition in Virtual Fitting

Hand gesture interaction has already become a research focus with the rapid development of virtual reality technologies. In this paper, we present a real-time interactive system based on hand gesture recognition in virtual fitting. Firstly, real-time data of hand gesture are captured by Leap Motion sensor. Then, we analyze the data and set the parameter projection to the hand model in virtual environment. Finally, the interaction with objects in the virtual environment will be realized by real-time gesture recognition. Application environment is designed as a virtual fitting room. Users can view virtual clothing by gesture interaction. Experiments show that the system is efficient and user friendly. And system’s gesture recognition has good real-time performance and accuracy.

A Virtual Music Control System Based on Dynamic Hand Gesture Recognition

Gesture Recognition technology has been widely used in virtual reality and human-computer interaction. This paper proposed a virtual music control system based on dynamic hand gesture recognition. The system mode was mainly designed and realized by three modules including control terminal, client terminal and the server. By capturing the gesture image sequence via a cellphone camera, the system is able to recognize information characters of gestures such as number of fingers and movement of gesture trace. Control terminal generate different instructions and send them to client terminal via server. Relative experiments showed that the interaction system had good applicability and portability.