Linear Discriminative Star Coordinates for Exploring Class and Cluster Separation of High Dimensional Data

One main task for domain experts in analysing their nD data is to detect and interpret class/cluster separations and outliers. In fact, an important question is, which features/dimensions separate classes best or allow a cluster‐based data classification. Common approaches rely on projections from nD to 2D, which comes with some challenges, such as: The space of projection contains an infinite number of items. How to find the right one? The projection approaches suffers from distortions and misleading effects. How to rely to the projected class/cluster separation? The projections involve the complete set of dimensions/features. How to identify irrelevant dimensions? Thus, to address these challenges, we introduce a visual analytics concept for the feature selection based on linear discriminative star coordinates (DSC), which generate optimal cluster separating views in a linear sense for both labeled and unlabeled data. This way the user is able to explore how each dimension contributes to clustering. To support to explore relations between clusters and data dimensions, we provide a set of cluster‐aware interactions allowing to smartly iterate through subspaces of both records and features in a guided manner. We demonstrate our features selection approach for optimal cluster/class separation analysis with a couple of experiments on real‐life benchmark high‐dimensional data sets.     

Minimum‐Displacement Overlap Removal for Geo‐referenced Data Visualization

Given a set of rectangles embedded in the plane, we consider the problem of adjusting the layout to remove all overlap while preserving the orthogonal order of the rectangles. The objective is to minimize the displacement of the rectangles. We call this problem Minimum -Displacement Overlap Removal (mdor ). Our interest in this problem is motivated by the application of displaying metadata of archaeological sites. Because most existing overlap removal algorithms are not designed to minimize displacement while preserving orthogonal order, we present and compare several approaches which are tailored to our particular usecase. We introduce a new overlap removal heuristic which we call re Arrange . Although conceptually simple, it is very effective in removing the overlap while keeping the displacement small. Furthermore, we propose an additional procedure to repair the orthogonal order after every iteration, with which we extend both our new heuristic and PRISM, a widely used overlap removal algorithm. We compare the performance of both approaches with and without this order repair method. The experimental results indicate that re Arrange is very effective for heterogeneous input data where the overlap is concentrated in few dense regions.     

A note to the integrable discretization of the mKdV and Schrödinger equations

An approach has been proposed to the integrable discretization of nonlinear evolution equations. Based on the bilinear formalism, we choose appropriate substitution from hyperbolic operator into continuous Hirota operators and obtain several new kinds of integrable system through seeking their 3-soliton solutions, such as the mKdV equation, the nonlinear Schrödinger equation and so on. By applying Adomian decompose method, we discuss the numerical analysis property to the discrete mKdV equation. In addition, we also point out the relations between the above discreted equations and some well-known equations.     

Panorama light‐field imaging

We present a novel approach to recording and computing panorama light fields. In contrast to previous methods that estimate panorama light fields from focal stacks or naive multi‐perspective image stitching, our approach is the first that processes ray entries directly and does not require depth reconstruction or matching of image features. Arbitrarily complex scenes can therefore be captured while preserving correct occlusion boundaries, anisotropic reflections, refractions, and other light effects that go beyond diffuse reflections of Lambertian surfaces.     

On‐the‐fly generation and rendering of infinite cities on the GPU

In this paper, we present a new approach for shape‐grammar‐based generation and rendering of huge cities in real‐time on the graphics processing unit (GPU). Traditional approaches rely on evaluating a shape grammar and storing the geometry produced as a preprocessing step. During rendering, the pregenerated data is then streamed to the GPU. By interweaving generation and rendering, we overcome the problems and limitations of streaming pregenerated data. Using our methods of visibility pruning and adaptive level of detail, we are able to dynamically generate only the geometry needed to render the current view in real‐time directly on the GPU. We also present a robust and efficient way to dynamically update a scene's derivation tree and geometry, enabling us to exploit frame‐to‐frame coherence. Our combined generation and rendering is significantly faster than all previous work. For detailed scenes, we are capable of generating geometry more rapidly than even just copying pregenerated data from main memory, enabling us to render cities with thousands of buildings at up to 100 frames per second, even with the camera moving at supersonic speed.     

Bilinear Accelerated Filter Approximation

Our method approximates exact texture filtering for arbitrary scales and translations of an image while taking into account the performance characteristics of modern GPUs. Our algorithm is fast because it accesses textures with a high degree of spatial locality. Using bilinear samples guarantees that the texels we read are in a regular pattern and that we use a hardware accelerated path. We control the texel weights by manipulating the u, v parameters of each sample and the blend factor between the samples. Our method is similar in quality to Cardinality‐Constrained Texture Filtering [ MS13 ] but runs two times faster.     

Feature‐Preserving Surface Completion Using Four Points

We present a user‐guided, semi‐automatic approach to completing large holes in a mesh. The reconstruction of the missing features in such holes is usually ambiguous. Thus, unsupervised methods may produce unsatisfactory results. To overcome this problem, we let the user indicate constraints by providing merely four points per important feature curve on the mesh. Our algorithm regards this input as an indication of an important broken feature curve. Our completion is formulated as a global energy minimization problem, with user‐defined spatial‐coherence constraints, allows for completion that adheres to the existing features. We demonstrate the method on example problems that are not handled satisfactorily by fully automatic methods.     

3D Motion Completion in Crowded Scenes

Crowded motions refer to multiple objects moving around and interacting such as crowds, pedestrians and etc. We capture crowded scenes using a depth scanner at video frame rates. Thus, our input is a set of depth frames which sample the scene over time. Processing such data is challenging as it is highly unorganized, with large spatio‐temporal holes due to many occlusions. As no correspondence is given, locally tracking 3D points across frames is hard due to noise and missing regions. Furthermore global segmentation and motion completion in presence of large occlusions is ambiguous and hard to predict. Our algorithm utilizes Gestalt principles of common fate and good continuity to compute motion tracking and completion respectively. Our technique does not assume any pre‐given markers or motion template priors. Our key‐idea is to reduce the motion completion problem to a 1D curve fitting and matching problem which can be solved efficiently using a global optimization scheme. We demonstrate our segmentation and completion method on a variety of synthetic and real world crowded scanned scenes.     

3D Shape Segmentation and Labeling via Extreme Learning Machine

We propose a fast method for 3D shape segmentation and labeling via Extreme Learning Machine (ELM). Given a set of example shapes with labeled segmentation, we train an ELM classifier and use it to produce initial segmentation for test shapes. Based on the initial segmentation, we compute the final smooth segmentation through a graph‐cut optimization constrained by the super‐face boundaries obtained by over‐segmentation and the active contours computed from ELM segmentation. Experimental results show that our method achieves comparable results against the state‐of‐the‐arts, but reduces the training time by approximately two orders of magnitude, both for face‐level and super‐face‐level, making it scale well for large datasets. Based on such notable improvement, we demonstrate the application of our method for fast online sequential learning for 3D shape segmentation at face level, as well as realtime sequential learning at super‐face level.