Yazhou Liu

Nonparametric background generation

A novel background generation method based on non-parametric background model is presented for background subtraction. We introduce a new model, named as effect components description (ECD), to model the variation of the background, by which we can relate the best estimate of the background to the modes (local maxima) of the underlying distribution. Based on ECD, an effective background generation method, most reliable background mode (MRBM), is developed. The basic computational module of the method is an old pattern recognition procedure, the mean shift, which can be used recursively to find the nearest stationary point of the underlying density function. The advantages of this method are three-fold: first, backgrounds can be generated from image sequence with cluttered moving objects; second, backgrounds are very clear without blur effect; third, it is robust to noise and small vibration. Extensive experimental results illustrate its good performance

3D Haar-Like Features for Pedestrian Detection

One basic observation for pedestrian detection in video sequences is that both appearance and motion information are important to model the moving people. Based on this observation, we propose a new kind of features, 3D Haar-like (3DHaar) features. Motivated by the success of Haar-like features in image based face detection and differential-frame based pedestrian detection, we naturally extend this feature by defining seven types of volume filters in 3D space, instead of using rectangle filter in 2D space. The advantage is that it can not only represent pedestrians appearance, but also capture the motion information. To validate the effectiveness of the proposed method, we combine the 3DHaar with support vector machine (SVM) for pedestrian detection. Our experiments demonstrate the 3DHaar are more effective for video based pedestrian detection.

An efficient coding scheme for surveillance videos captured by stationary cameras

In this paper, a new scheme is presented to improve the coding efficiency of sequences captured by stationary cameras (or namely, static cameras) for video surveillance applications. We introduce two novel kinds of frames (namely background frame and difference frame) for input frames to represent the foreground/background without object detection, tracking or segmentation. The background frame is built using a background modeling procedure and periodically updated while encoding. The difference frame is calculated using the input frame and the background frame. A sequence structure is proposed to generate high quality background frames and efficiently code difference frames without delay, and then surveillance videos can be easily compressed by encoding the background frames and difference frames in a traditional manner. In practice, the H.264/AVC encoder JM 16.0 is employed as a build-in coding module to encode those frames. Experimental results on eight in-door and out-door surveillance videos show that the proposed scheme achieves 0.12 dB~1.53 dB gain in PSNR over the JM 16.0 anchor specially configured for surveillance videos.

Nonparametric Background Generation

A novel background generation method based on non-parametric background model is presented for background subtraction. We introduce a new model, named as effect components description (ECD), to model the variation of the background, by which we can relate the best estimate of the background to the modes (local maxima) of the underlying distribution. Based on ECD, an effective background generation method, most reliable background mode (MRBM), is developed. The basic computational module of the method is an old pattern recognition procedure, the mean shift, which can be used recursively to find the nearest stationary point of the underlying density function. The advantages of this method are three-fold: first, backgrounds can be generated from image sequence with cluttered moving objects; second, backgrounds are very clear without blur effect; third, it is robust to noise and small vibration. Extensive experimental results illustrate its good performance

Granularity-tunable gradients partition (GGP) descriptors for human detection

This paper proposes a novel descriptor, granularity-tunable gradients partition (GGP), for human detection. The concept granularity is used to define the spatial and angular uncertainty of the line segments in the Hough space. Then this uncertainty is backprojected into the image space by orientation-space partitioning to achieve efficient implementation. By changing the granularity parameter, the level of uncertainty can be controlled quantitatively. Therefore a family of descriptors with versatile representation property can be generated. Specifically, the finely granular GGP descriptors can represent the specific geometry information of the object (the same as Edgelet); while the coarsely granular GGP descriptors can provide the statistical representation of the object (the same as histograms of oriented gradients, HOG). Moreover, the position, orientation, strength and distribution of the gradients are embedded into a unified descriptor to further improve the GGPs representation power. A cascade structured classifier is built by boosting the linear regression functions. Experimental results on INRIA dataset show that the proposed method achieves comparable results to those of the state-of-the-art methods.

Contour-motion feature (CMF): A space-time approach for robust pedestrian detection

This paper presents a contour-motion feature for robust pedestrian detection. The space-time contours are used as the low level representation of the pedestrian. Then we apply 3D distance transform to extend the 1-dimensional contour into 3-dimensional space. By this way, the relations between the local contours can be maintained implicitly. Further, by encapsulating the static and dynamic information by 3D Haar-like filters, we can generate the middle level pedestrian representation: contour-motion features. Then we use boosting method to select the most representative features. Our experiments demonstrate that the proposed approach can outperform Violas well-known pedestrian detector in both detection accuracy and generalization ability. In addition, even though our approach is presented in pedestrian detection scenario, it has been extended to human activity recognition application and remarkable performance has been achieved.

Silicone oil emulsions: strategies to improve their stability and applications in hair care products.

Silicone oils have wide range of applications in personal care products due to their unique properties of high lubricity, non‐toxicity, excessive spreading and film formation. They are usually employed in the form of emulsions due to their inert nature. Until now, different conventional emulsification techniques have been developed and applied to prepare silicone oil emulsions. The size and uniformity of emulsions showed important influence on stability of droplets, which further affect the application performance. Therefore, various strategies were developed to improve the stability as well as application performance of silicone oil emulsions. In this review, we highlight different factors influencing the stability of silicone oil emulsions and explain various strategies to overcome the stability problems. In addition, the silicone deposition on the surface of hair substrates and different approaches to increase their deposition are also discussed in detail.

Image evaluation factors

We describe a method for objective and quantitative evaluation of image quality. The method represents a novel use of image enhancement concepts. It employs three new measures that evaluate the definition of contours, uniform intensity distribution, and noise rate in determining the image quality. Because the three measures have clear physical meanings, they can be selectively applied according to the viewers evaluation criteria. The three measures are relatively inexpensive to compute, making them suitable for automated ranking of image quality in personal digital imaging devices, such as digital cameras. However, the method is equally adept at evaluating other digital images such as those on the Internet. Experiments with the method show good correlation with visual quality assessment for various image subject types.

A texture-based tamper detection scheme by fragile watermark

A texture-based tamper detection scheme by a fragile watermarking technique is proposed in this paper. In comparison with other fragile watermarking schemes, the highlight of our scheme is that it is rather sensitive to malicious tamper such as replacing ones face in the image by anothers and at the same time it is insensitive to other legal processing such as lossy JPEG compression and brightness/contrast changes. So it is more suitable for tamper detection in practical use.

Single sequential minimal optimization: an improved SVMs training algorithm

We introduce homogeneous coordinates to represent support vector machines (SVMs) and develop a corresponding training algorithm: single sequential minimal optimization (SSMO). By this simple trick (homogeneous coordinates representation), linear constrains will not appear in quadratic programming (QP) optimization problem. So unlike the most popular used SVM training algorithm sequential minimal optimization (SMO) which solves the QP subproblem containing minimal two Lagrange multipliers, SSMO can analytically update only one Lagrange multiplier at every step. Because of avoiding double loops in heuristically choosing the two Lagrange multipliers in SMO, both CPU time and iterations can be decreased greatly. Experiments on MNIST database, under mild KKT conditions accuracy requirement, shows SSMO can be more than 2 times faster than SMO.