细分曲面求交交线计算方法的研究

主要针对三角网格的细分曲面求交提出了一种有效的交线计算的方法,该方法适用于任意三角网格的细分曲面中.在利用AABB和二部图进行初始控制网格相交性检测后,利用该方法快速有效地求出细分曲面的交线.     

基于图像建模的多视图模型合并

通过基于单幅图像的建模方法或者基于立体像对的建模方法,可以重建图像序列中每一幅图像或者立体图像对对应的模型。但所得的是局部模型,并非物体的完整模型。针对这一问题,提出了一种基于序列图像建模的多视图模型合并方法。基于已重建出的同一物体不同角度的三角形网格模型,在统一坐标系下,对相邻角度下的三角形网格模型求取其重合区域,记录刚好包围重合区域的原始网格区域,对该区域重新进行网格划分,最后得到物体完整的三角网格模型。实验结果表明,该方法是可行的、有效的。     

基于多流网络一致性的视频显著性检测

现有的视频显著性检测算法通常采用双流结构提取视频的时空线索,其中运动信息作为双流结构的一个分支,在显著物体发生剧烈或慢速移动时存在运动估计准确率低的问题,并且不合理的训练数据或方案使得权重偏向单个分支结构。提出一种基于多流网络一致性的视频显著性检测算法MSNC。设计并使用一种新的三重网络结构提取预选目标区域的颜色信息、时序信息和先验特征,通过先验特征补偿运动流的缺陷,并提高运动线索的利用率。采用多流一致性融合模型优化三流分支,得到不同特征的最佳融合方案。同时通过循环训练策略平衡三重网络的权重,以避免网络过度拟合单流分支,从而有效地提高运动估计和定位的准确率。在Davis数据集上的实验结果表明,相比PCSA、SSAV、MGA等算法,该算法的鲁棒性更优,其maxF和S-Measure值分别达到0.893和0.912,MAE仅为0.021。     

结合l1/2范数与显著性约束的背景减除

传统背景减除模型在背景静止和前景对象移动较快时提取到的前景效果较好,但当背景变化或前景对象移动缓慢时容易将动态背景误判为前景或检测出的前景有较多空洞。针对传统背景减除模型在动态背景和前景对象移动缓慢条件下存在前景检测精度下降的问题,提出一种基于l_1/2范数与显著性约束的背景减除模型。将观测数据分为低秩背景、运动前景和动态干扰3类,利用l_1/2范数约束运动前景加强前景稀疏性,有效抑制动态背景对前景提取造成的干扰,提高运动前景在动态背景中的检测精度。引入视频每一帧的显著性约束,通过对每一帧图像进行低秩稀疏分解来检测移动缓慢的目标。实验结果表明,该模型对于复杂场景具有较强的适应能力,可有效去除动态背景对前景的干扰,快速检测出移动缓慢的前景对象,相比于l_1/1/2-RPCA背景减除模型的平均查全率、查准率和调和平均值分别提升了9、14和10个百分点。     

Erosion coupled dynamic mesh analysis for location of maximum erosion in 90-degree bends for gas-solid flows

The progressive change in the surface geometry of the component due to erosive wear affects the correct estimation of erosive wear performance and service life of the components handling particulate flows. The current study focuses on determining the change in the location of higher erosion on the bend surface during the pneumatic conveying of solids with continuous geometric modification due to erosive wear. Computational fluid dynamics (CFD) based erosion-coupled dynamic mesh methodology is adopted to simulate the time-dependent surface modification of the 90° bend geometry due to erosive wear. Available experimental data are used to validate the numerical results. Further, the erosion distribution and the location of the maximum erosion for different flow velocities, particle sizes, and bend radius ratios with the increase in solid throughput are investigated. It has been found that the modification in the bend geometry due to erosion influences the location of the maximum erosion. The increase in thickness loss due to erosion increases the variation in the location of the maximum erosion. Furthermore, an equation for predicting the location of maximum erosion of bend geometry is obtained based on the bend radius ratio and the thickness loss.     

Efficient random saliency map detection

Most image retargeting algorithms rely heavily on valid saliency map detection to proceed. However, the inefficiency of high quality saliency map detection severely restricts the application of these image retargeting methods. In this paper, we propose a random algorithm for efficient context-aware saliency map detection. Our method is a multiple level saliency map detection algorithm that integrates multiple level coarse saliency maps into the resulting saliency map and selectively updates unreliable regio...     

Adaptive triangular-mesh reconstruction by mean-curvature-based refinement from point clouds using a moving parabolic approximation

We use a moving parabolic approximation (MPA) to reconstruct a triangular mesh that approximates the underlying surface of a point cloud from closed objects. First, an efficient strategy is presented for constructing a hierarchical grid with adaptive resolution and generating an initial mesh from point clouds. By implementing the MPA algorithm, we can estimate the differential quantities of the underlying surface, and subsequently, we can obtain the local quadratic approximants of the squared distance function for any point in the vicinity of the target shape. Thus, second, we adapt the mesh to the target shape by an optimization procedure that minimizes a quadratic function at each step. With the objective of determining the geometrical features of the target surface, we refine the approximating mesh selectively for the non-flat regions by comparing the estimated curvature from the point clouds and the estimated curvatures computed from the current mesh. Finally, we present various examples that demonstrate the robustness of our method and show that the resulting reconstructions preserve geometric details.     

Numerical simulation of detonation using an adaptive Cartesian cut-cell method combined with a cell-merging technique

A second-order accurate scheme for the Cartesian cut-cell method developed previously by the authors [Ji H, Lien F-S, Yee E. Comput. Methods Appl. Mech. Eng. 198 (2008), 432] is generalized for application to both two- and three-dimensional inviscid compressible flow problems. A cell-merging approach is used to address the so-called “small cell” problem that has plagued Cartesian cut-cell methods. In the present cell-merging approach, the conservative variables are stored at the cut-cell centroid (including the non-merged and merged cut-cells) rather than at the Cartesian cell center. Although this approach results in a more complicated search algorithm for the determination of the neighbor cells (required for the computation of the spatial gradients of the conservative variables), this approach enables the straightforward formulation of a higher than first-order accurate discretization scheme in the vicinity of the (complex and irregular) internal boundaries of the flow domain. Six test cases (including detonation problems) are used to demonstrate the accuracy and capability of the adaptive cut-cell method, for which both mesh refinement and derefinement techniques are employed in the case of an unsteady shock diffraction problem.     

Using a Visual Attention Model to Improve Gaze Tracking Systems in Interactive 3D Applications

This paper introduces the use of a visual attention model to improve the accuracy of gaze tracking systems. Visual attention models simulate the selective attention part of the human visual system. For instance, in a bottom‐up approach, a saliency map is defined for the image and gives an attention weight to every pixel of the image as a function of its colour, edge or intensity. Our algorithm uses an uncertainty window, defined by the gaze tracker accuracy, and located around the gaze point given by the tracker. Then, using a visual attention model, it searches for the most salient points, or objects, located inside this uncertainty window, and determines a novel, and hopefully, better gaze point. This combination of a gaze tracker together with a visual attention model is considered as the main contribution of the paper. We demonstrate the promising results of our method by presenting two experiments conducted in two different contexts: (1) a free exploration of a visually rich 3D virtual environment without a specific task, and (2) a video game based on gaze tracking involving a selection task. Our approach can be used to improve real‐time gaze tracking systems in many interactive 3D applications such as video games or virtual reality applications. The use of a visual attention model can be adapted to any gaze tracker and the visual attention model can also be adapted to the application in which it is used.     

SPC: Fast and Efficient Scalable Predictive Coding of Animated Meshes

Animated meshes are often represented by a sequence of static meshes with constant connectivity. Due to their frame-based representation they usually occupy a vast amount of bandwidth or disk space. We present a fast and efficient scalable predictive coding (SPC) scheme for frame-based representations of animated meshes. SPC decomposes animated meshes in spatial and temporal layers which are efficiently encoded in one pass through the animation. Coding is performed in a streamable and scalable fashion. Dependencies between neighbouring spatial and temporal layers are predictively exploited using the already encoded spatio-temporal neighbourhood. Prediction is performed in the space of rotation-invariant coordinates compensating local rigid motion. SPC supports spatial and temporal scalability, and it enables efficient compression as well as fast encoding and decoding. Parts of SPC were adopted in the MPEG-4 FAMC standard. However, SPC significantly outperforms the streaming mode of FAMC with coding gains of over 33%, while in comparison to the scalable FAMC, SPC achieves coding gains of up to 15%. SPC has the additional advantage over FAMC of achieving real-time encoding and decoding rates while having only low memory requirements. Compared to some other non-scalable state-of-the-art approaches, SPC shows superior compression performance with gains of over 16% in bit-rate.