PG2015short

Permanent URI for this collection

https://diglib.eg.org/handle/10.2312/13267

Browse

Now showing 1 - 15 of 15

Accelerating Graph-based Path Planning Through Waypoint Clustering
(The Eurographics Association, 2015) Wardhana, Nicholas Mario; Johan, Henry; Seah, Hock-Soon; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Modern Computer Graphics applications commonly feature very large virtual environments and diverse characters which perform different kinds of motions. To accelerate path planning in such scenario, we propose subregion graph data structure. It consists of subregions, which are clusters of locally connected waypoints inside a region, as well as their connectivities. We also present a fast algorithm to automatically generate subregion graph from enhanced waypoint graph map representation, which also supports various motion types and can be created from large virtual environments. Nevertheless, subregion graph can also be generated from any graph-based map representation. Our experiments showed that subregion graph is very compact relative to the input waypoint graph. By firstly planning subregion path, and then limiting waypoint-level planning to the subregion path, up to 8 times average speedup can be achieved, while average length ratios are maintained at as low as 102.5%.
Simultaneous Stereo Rectification and Distortion Correction with Application to DoF Synthesis
(The Eurographics Association, 2015) Yeh, Chen Ting; Ho, Tien-Yu; Huang, Szu-Hao; Lai, Shang-Hong; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Stereo rectification based on epipolar geometry is an image transformation process which is used to align the rectified stereo images to be coplanar. For image pairs captured by low-cost stereo camera systems, image noises and optical distortions in images may lead to errors in depth estimation. In this paper, a novel algorithm is proposed to simultaneously solve the stereo rectification and distortion correction problems in an integrated optimization framework. The estimated disparity maps are then used to synthesize the depth of field (DoF), which is to increase the saliency of the foreground objects, with the proposed spatially-variant filtering method. Experimental results on various synthetic and real stereo images are shown to demonstrate that the proposed algorithm can effectively reduce the stereo rectification error by considering the optical distortion model. The DoF synthesis can also be improved by using more accurate disparity estimation results.
Frontmatter: Pacific Graphics 2015 - Short Papers
(The Eurographics Association, 2015) Stam, Jos; Mitra, Niloy J.; Xu, Kun; -
A Voronoi based Labeling Approach to Curve Reconstruction and Medial Axis Approximation
(The Eurographics Association, 2015) Peethambaran, Jiju; Parakkat, Amal Dev; Muthuganapathy, Ramanathan; Stam, Jos and Mitra, Niloy J. and Xu, Kun
In this paper, we present a Voronoi based algorithm for closed curve reconstruction and medial axis approximation from planar points. In principle, the algorithm estimates one of the poles (farthest Voronoi vertices of a Voronoi cell) and hence the normals at each sample point by drawing an analogy between a residential water distribution system and Voronoi diagram of input samples. The algorithm then labels Voronoi vertices as either inner or outer with respect to the original curve and subsequently construct a piece-wise linear approximation to the boundary and the interior medial axis of the original curve for a class of curves having bi-tangent neighborhood convergence (BNC). The proposed algorithm has been evaluated for its usefulness using various test data. Results indicate that, even sparsely and non-uniformly sampled curves with sharp corners, outliers or collection of curves are faithfully reconstructed by the proposed algorithm.
Light-Guided Tree Modeling of Diverse Biomorphs
(The Eurographics Association, 2015) Yi, Lei; Li, Hongjun; Guo, Jianwei; Deussen, Oliver; Zhang, Xiaopeng; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Creation of tree models faithful to light environment is an important task in computer graphics as well as in botanical research, such as horticulture and forestry. In this paper, we propose an approach to model virtual trees with constraints of light resources and tree morphological properties. By the allocation of received light resources, tree model parameters are estimated, including branching directions and branching sizes. The light energy is calculated by sampling the environmental space, so that the final architecture of trees could be modeled corresponding to its growth environments. Experimental results show that the proposed method create trees with architectures guided by light resources.
Superpixels Generation of RGB-D Images Based on Geodesic Distance
(The Eurographics Association, 2015) Pan, Xiao; Zhou, Yuanfeng; Liu, Shuwei; Zhang, Caiming; Stam, Jos and Mitra, Niloy J. and Xu, Kun
A novel algorithm for generating superpixels of RGB-D images is presented in this paper. A regular triangular mesh is constructed by the depth and a local geometric features sensitive initialization method is proposed for initializing seeds by a density function. Over-segmentation of the vertices on mesh can be generated by minimizing a new energy function defined by weighted geodesic distance which can be used for measuring the similarity of vertices with color information. At last, superpixels are generated by re-mapping the mesh over-segmentation to 2D image. During energy optimizing, we will check the topology correctness of the superpixels and refine the topology of the superpixels. Experiments on a large RGB-D images database show that the superpixels generated by the new method can adhere to the object boundaries well and outperform the state-of-the-art methods.
Incomplete 3D Shape Retrieval via Sparse Dictionary Learning
(The Eurographics Association, 2015) Wan, Lili; Jiang, Jingyu; Zhang, Hao; Stam, Jos and Mitra, Niloy J. and Xu, Kun
How to deal with missing data is one of the recurring questions in data analysis. The handling of significant missing data is a challenge. In this paper, we are interested in the problem of 3D shape retrieval where the query shape is incomplete with moderate to significant portions of the original shape missing. The key idea of our method is to grasp the basis local descriptors for each shape in the retrieved database by sparse dictionary learning and apply them in sparsely coding the local descriptors of an incomplete query. First, we present a method of computing heat kernel signatures for incomplete shapes. Next, for each shape in the database, a set of basis local descriptors, which is called a dictionary, is learned and taken as its representative. Finally, a query incomplete shape's heat kernel signatures are respectively reconstructed by each dictionary, and the shape similarities are therefore measured by the reconstruction errors. Experimental results show that the proposed method has achieved significant improvements over previous works on retrieving non-rigid incomplete shapes.
Adaptive Hierarchical Shape Matching
(The Eurographics Association, 2015) Tian, Yuan; Yang, Yin; Guo, Xiaohu; Prabhakaran, Balakrishnan; Stam, Jos and Mitra, Niloy J. and Xu, Kun
In this paper, we present an adaptive hierarchical method allowing users to interact with geometrically complex 3D deformable objects based on an extended shape matching approach. Our method extends the existing multiresolution shape matching methods with improved energy convergence rate. This is achieved by using adaptive integration strategies to avoid insignificant shape matching iterations during the simulation. As demonstrated in our experimental results, the proposed method provides an efficient yet stable deformable simulation of complex models in real-time.
Modal Space Subdivision for Physically-plausible 4D Shape Sequence Completion from Sparse Samples
(The Eurographics Association, 2015) Xia, Qing; Li, Shuai; Qin, Hong; Hao, Aimin; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Subdivision techniques are powerful and ubiquitous in generating smooth surfaces from coarse 3D polygonal models, yet space-time subdivision remains unexplored when the goal is to produce dense physically-realistic shape sequence with smooth transition from a set of sparse 3D model samples. In this paper, we considers a completely data-driven strategy and proposes a novel modal space subdivision scheme to facilitate the rapid generation of physically-plausible "in-between" shapes from sparse model samples. The key idea of our approach is to abstract any shape as a point in modal space, represent its potential deformation with shape modes, and iteratively seek all the intermediate shapes between two consecutive models in the modal space via recursive subdivision. Meanwhile, to guarantee deformation details, we further interpolate the shape in the local regions after global modal-space subdivision. Comprehensive experiments on various model inputs have demonstrated the power, versatility, high performance, and potential of our modal space subdivision scheme.
Pairwise Surface Registration Using Local Voxelizer
(The Eurographics Association, 2015) Song, Peng; Chen, Xiaoping; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Surface registration is the process that brings scans into a common coordinate system by aligning their overlapping components, which can be achieved by finding a few pairs of matched points on each scan pair using shape descriptors and employing the matches to compute an alignment transformation. This paper proposes a local voxelizer descriptor, and the key idea is to define a unique local reference frame (LRF) using the local shape around a basis point, perform voxlization for the local shape within a cubical volume aligned with the LRF, and concatenate local features extracted from each voxel to construct the descriptor. A pairwise registration algorithm is developed by choosing a single pair of matched points using the local voxelizer descriptor, and computing a rigid transformation based on aligning the corresponding LRFs. Quantitative experiments show that our algorithm can register scan pairs with small overlap, while maintaining acceptable registration accuracy.
Robust and GPU-friendly Isotropic Meshing Based on Narrow-banded Euclidean Distance Transformation
(The Eurographics Association, 2015) Leung, Yuen Shan; Wang, Xiaoning; He, Ying; Liu, Yong-Jin; Wang, Charlie C. L.; Stam, Jos and Mitra, Niloy J. and Xu, Kun
In this paper, we propose a simple-yet-effective method for isotropic meshing via Euclidean distance transformation based Centroidal Voronoi Tessellation (CVT). The proposed approach aims at improving the performance as well as robustness of computing CVT on curved domains while simultaneously maintaining the high-quality of the output meshes. In contrast to the conventional extrinsic methods which compute CVTs in the entire volume bounded by the input model, our idea is to restrict the computation in a 3D shell space with user-controlled thickness. Taking the voxels which contain the surface samples as the sites, we compute the exact Euclidean distance transform on the GPU. Our algorithm is fully parallel and memory-efficient, and it can construct the shell space with resolution up to 2048<sup>3</sup> at interactive speed. Since the shell space is able to bridge holes and gaps up to a certain tolerance, and tolerate non-manifold edges and degenerate triangles, our algorithm works well on models with such defects, whereas the conventional remeshing methods often fail.
Structure-Preserving Image Smoothing via Phase Congruency-aware Weighted Least Square
(The Eurographics Association, 2015) Yu, Jinze; Sato, Yoichi; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Structure-preserving image smoothing, or also understood as structure-texture separation problem, is an important topic for both computer vision and computer graphics as structure-texture separation can help better image understanding. In fact, many image processing problems can be well achieved once two layers possessing different properties of a scene are separated. Therefore better separating structure and texture from an image is of great practical importance. However, it is also a challenge topic since it is often quite subjective to tell the difference between the two layers. Recently, researchers made great efforts on separating a given image into its structure and texture layers by distinguishing edges from oscillations based on non-gradients-based descriptors or descriptors defined specifically for certain kinds of image data. These methods show advantages compared to the purely gradients-based methods with extra information provided besides gradients. In this paper, we propose a structure-texture separation method using non-gradients-based descriptor. Specially, we propose an alternative yet simple image smoothing approach based on the well-known weighted least square (WLS) framework. Our approach combines the phase congruency features that can better help locate structure or contour information of objects. Phase congruency performs well for distinguishing the structure and texture as it mimics the response of the human perception system to contours and is also sensitive to periodic patterns. By including the phase congruency as weights, WLS can better smooth out images while preserving structures. Experimental results indicate that the proposed approach is effective for structure-texture separation and achieves low computational complexity, compared to the state-of-the-art methods.
Complex Modulation Computer-Generated Hologram with Occlusion Effect by a Fast Hybrid Point-source/Wave-field Approach
(The Eurographics Association, 2015) Gilles, Antonin; Gioia, Patrick; Cozot, Rémi; Morin, Luce; Stam, Jos and Mitra, Niloy J. and Xu, Kun
We propose a fast Computer-Generated Hologram (CGH) computation method with occlusion effect based on a hybrid point-source/wave-field approach. Whereas previously proposed methods tried to reduce the computational complexity of the point-source or the wave-field approaches independently, the proposed method uses the two approaches together and therefore takes advantages from both of them. Our algorithm consists of three steps. First, the 3D scene is sliced into several depth layers parallel to the hologram plane. Then, light scattered by the scene is propagated and shielded from one layer to another, starting from the farthest layer. For each layer, light propagation and light shielding are performed using either a point-source or a wave-field approach according to a threshold criterion on the number of points within the layer. Finally, we compute light propagation from the nearest layer to the hologram plane in order to obtain the final CGH. Experimental results reveal that this combination of approaches does not produce any visible artifact and outperforms both the point-source and wavefield approaches.
Efficient Interactive Image Segmentation with Local and Global Consistency
(The Eurographics Association, 2015) Li, Hong; Wu, Wen; Wu, Enhua; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Interactive image segmentation models aim to classify the image pixels into foreground and background classes given some foreground and background scribbles. In this paper, we propose a novel framework for interactive image segmentation which builds upon the local and global consistency model. The final segmentation results are improved by tackling two disadvantages in graph construction of traditional models: graph structure modeling and graph edge weights formation. The scribbles provided by users are treated as the must-link and must-not-link constraints. Then the graph structure is modeled as an approximately k-regular sparse graph by integrating these constraints and our extended neighboring spatial relationships. Content driven locally adaptive kernel parameter is proposed to tackle the insufficiency of previous models which usually employ a unified kernel parameter. After the graph construction, a novel three-stage strategy is proposed to get the final segmentation results. Experimental results and comparisons with other state-of-the-art methods demonstrate that our framework can efficiently and accurately extract foreground objects from background.
Parallel Importing of OBJ Meshes in CUDA
(The Eurographics Association, 2015) Possemiers, Aidan. L.; Lee, Ickjai; Stam, Jos and Mitra, Niloy J. and Xu, Kun
Alias | Wavefront OBJ meshes are a common text file type for transferring 3D mesh data between applications made by different vendors. However, as the mesh complexity gets higher and denser, the files become larger and slower to import. This paper explores the use of GPUs to accelerate the importing and parsing of OBJ files by studying file read-times, runtimes and load resistance. We propose a new method of reading and parsing that circumvents GPU architecture limitations and improves performance, seeing the new GPU method outperform CPU methods with a 6-8x speedup. When run on a heavily loaded system, the new method only received an 80% performance hit, compared to the 160% that the CPU methods received. The loaded GPU speedup compared to unloaded CPU methods was 3.5x, and, when compared to loaded CPU methods, 8x. These results demonstrate that the time is right for further research into the use of data-parallel GPU acceleration beyond that of computer graphics and high performance computing.

Browse

Browsing PG2015short by Issue Date

Results Per Page

Sort Options