Abstract

Global optimization of MRF energy using graph cuts iswidely used in computer vision. As the images are gettinglarger, faster graph cuts are needed without sacrificing op-timality. Initializing or reparameterizing a graph using re-sults of a similar one has provided efficiency in the past. Inthis paper, we present a method to speedup graph cuts us-ing shrink-expand reparameterization. Our scheme mergesthe nodes of a given graph to shrink it. The resulting graphand its mincut are expanded and used to reparameterize theoriginal graph for faster convergence. Graph shrinking canbe done in different ways. We use a block-wise shrinkingsimilar to multiresolution processing of images in our Mul-tiresolution Cuts algorithm. We also develop a hybrid ap-proach that can mix nodes from different levels without af-fecting optimality. Our algorithm is particularly suited forprocessing large images. The processing time on the fulldetail graph reduces nearly by a factor of 4. The overallapplication time including all book-keeping is faster by afactor of 2 on various types of images

Abstract

Displays remain flat and passive amidst the many changes in their fundamental technologies. One natural step ahead is tocreate displays that merge seamlessly in shape and appearance with one’s natural surroundings. In this paper, we present a system todesign, render to, and build view-dependentmultiplanar displaysof arbitrary piecewise-planar shapes, built using polygonal facets.Our system provides high quality, interactive rendering of 3D environments to a head-tracked viewer on arbitrary multiplanar displays.We develop a novel rendering scheme that produces exact image and depth map at each facet, producing artifact-free images on andacross facet boundaries. The system scales to a large number of display facets by rendering all facets in a single pass of rasterization.This is achieved using a parallel, perframe, view-dependent binning and prewarping of scene triangles. The display is driven using oneor more targetquilt imagesinto which facet pixels are packed. Our method places no constraints on the scene or the display and allowsfor fully dynamic scenes to be rendered interactively at high resolutions. The steps of our system are implemented efficiently oncommodity GPUs. We present a few prototype displays to establish the scalability of our system on different display shapes, formfactors, and complexity: from a cube made out of LCD panels to spherical/cylindrical projected setups to arbitrary complex shapes insimulation. Performance of our system is demonstrated for both rendering quality and speed, for increasing scene and display facetsizes. A subjective user study is also presented to evaluate the user experience using a walk-around display compared to a flat panelfor a game-like setting.

Abstract

Matching patches of a source image with patches of itself ora target image is a first step for many operations. Finding the optimumnearest-neighbors of each patch using a global search of the image isexpensive. Optimality is often sacrificed for speed as a result. We presentthe Mixed-Resolution Patch-Matching (MRPM) algorithm that uses apyramid representation to perform fast global search. We compare mixed-resolution patches at coarser pyramid levels to alleviate the effects ofsmoothing. We store more matches at coarser resolutions to ensure widersearch ranges and better accuracy at finer levels. Our method achievesnear optimality in terms of average error compared to exhaustive search.Our approach is simple compared to complex trees or hash tables used byothers. This enables fast parallel implementations on the GPU, yieldingupto 70×speedup compared to other iterative approaches. Our approachis best suited when multiple, global matches are needed

Abstract

Large scale reconstructions of camera matrices and pointclouds have been created using structure from motion from communityphoto collections. Such a dataset is rich in information; it representsa sampling of the geometry and appearance of the underlying space.In this paper, we encode the visibility information between and amongpoints and cameras asvisibility probabilities. The conditional visibilityprobability of a set of points on a point (or a set of cameras on a camera)can rank points (or cameras) based on their mutual dependence. Wecombine the conditional probability with a distance measure to prioritizepoints for fast guided search for the image localization problem. Wedefine dual problem of feature triangulation as finding the 3D coordinatesof a given image feature point. We use conditional visibility probabilityto quickly identify a subset of cameras in which a feature is visible

Abstract

Graphics models are getting increasingly bulkier with de-tailed geometry, textures, normal maps, etc. There is a lotof interest to model and navigate through detailed models oflarge monuments. Many monuments of interest have bothrich detail and large spatial extent. Rendering them for navi-gation on a single workstation is practically impossible, evengiven the power of today’s CPUs and GPUs. Many modelsmay not fit the GPU memory, the CPU memory, or eventhe secondary storage of the CPU. Distributed renderingusing a cluster of workstations is the only way to navigatethrough such models. In this paper, we present a design ofa distributed rendering system intended for massive mod-els. Our design has a server that holds the skeleton of thewhole model, namely, its scenegraph with actual geometryreplaced by bounding boxes at all levels. The server dividesthe screen space among a number of clients and sends thema list of objects they need to render using a frustum cullingstep. The clients use 2 GPUs with one devoted to visibilityculling and the other to rendering. Frustum culling at theserver, visibility culling on one GPU, and rendering on thesecond GPU form the stages of our distributed renderingpipeline. We describe the design and implementation of oursystem and demonstrate the results of rendering relativelylarge models using different clusters of clients in this paper

Abstract

We present a scheme for interactive ray tracing of Bezier bicubicpatches using Newton iteration in this paper. We use a mixed hi-erarchy representation as the acceleration structure. This has abounding volume hierarchy above the patches and a fixed depthsubpatch tree below it. This helps reduce the number of ray-patchintersections that needs to be evaluated and provides good initial-ization for the iterative step, keeping the memory requirements low.We use Newton iteration on the generated list of ray patch intersec-tions in parallel. Our method can exploit the cores of the CPU andthe GPU with OpenMP on the CPU and CUDA on the GPU bysharing work between them according to their relative speeds. Adata parallel framework is used throughout starting with a list ofrays, which is transformed to a list of ray-patch intersections bytraversal and then to intersections and a list of secondary rays byroot finding. Shadow and reflection rays can be handled exactly inthe same manner as a result. We also show how our method ex-tends easily to generate soft shadows using area light sources andpath tracing by tracing a large number of rays per pixel. We rendera million pixel image of the Teapot model at 125 fps on a systemwith an Intel i7 920 and a Nvidia GTX580 for primary rays onlyand at about 65 fps with one pass of shadow and refection rays.

Abstract

Displays have seen much improvements over the years, withenhancements in spatial resolution and vertical refresh, etc.,to provide better and smoother visual experiences. Color in-tensity resolution, however, has not changed much over thepast few decades. Most displays are still limited to 8-bitsper channel. Simultaneously, much work has gone into cap-turing high dynamic range images. Mapping these directlyto current displays loses information that may be critical tomany applications. We present a way to enhance intensityresolution of a given display by mixing intensities over spa-tial or temporal domains. Our system sacrifices high verticalrefresh and spatial resolution in order to gain intensity res-olution. We present three ways to mix intensities: spatially,temporally and spatio-temporally. The systems produce in-between-intensities not present on the base display, whichare clearly distinguishable by the naked eye. We evaluateour systems using both a camera and human subjects, eval-uating whether they scale the intensity resolution and alsoensuring that the newly generated intensities follow the dis-play model

Abstract

Browsers of personal digital photographs all essentially fol-low the slide show paradigm, sequencing through the pho-tos in the order they are taken. A more engaging way tobrowse personal photographs, especially of a large space likea popular monument, should involve the geometric contextof the space. In this paper, we present ageometry directedphoto browserthat enables users to browse their personalpictures with the underlying geometry of the space to guidethe process. The browser uses a pre-computed package ofgeometric information about the monument for this task.The package is used to register a set of photographs takenby the user with the common geometric space of the monu-ment. This involves localizing the images to the monumentspace by computing the camera matrix corresponding to it.We use a state-of-the-art method for fast localization. Reg-istered photographs can be browsed using a visualizationmodule that shows them in the proper geometric contextwith respect to a point-based 3D model of the monument.We present the design of the geometry-directed browser anddemonstrate its utility for a few sequences of personal im-ages of well-known monuments. We believe personal photobrowsers can provide an enhanced sense of one’s own ex-perience with a monument using the underlying geometriccontext of the monument.

Abstract

: Density functional B3LYP method was used to investigate the preference of intra and intermolecular cyclizations of linear tripeptides containing tetrahydrofuran amino acids. Two distinct model pathways were conceived for the cyclization reaction and all possible transition states and intermediates were located. Analysis of the energetics indicate intermolecular cyclization being favored by both thermodynamic and kinetic control. Geometric and NBO analyses were performed to explain the trends obtained along both the reaction pathways. Conceptual density functional theory based reactive indices also show that reaction pathways leading to intermolecular cyclization of the tripeptides are relatively more facile compared to intramolecular cyclization.

Abstract

Theoretical justication for preferential heterochiral cyclic trimerization of 5- (aminoethyl)-2-furancarboxylic acid (AEFC) is attempted using density functional theory (DFT) calculations. Results from explicit solvent assisted reaction pathways indicate greater stability of heterochiral cyclic tripeptides over their homochiral counterparts, contrary to ndings from gas phase and implicit solvent phase results. Pathways explored at M06/6- 31G(d,p) and MP2/6-31G(d,p) levels of theory show kinetic preference for heterochiral cy- clization. Analysis of optimized geometries reveals existence of strong hydrogen bonding interactions in the solvated heterochiral tripeptides. Thus, the ability of the cyclic tripep- tides to form strong noncovalent interactions increases with conversion of stereochemistry at one of its chiral centers from homo to heterochiral conformation. The resulting change in molecular symmetry facilitates the interacting sites to reorient such that the peptide can in- teract with a nucleophile from both the faces. This is further substantiated by computed IR spectra, NBO and AIM data. Additionally, justication for the stability of heterochiral cyclic tripeptides comes from molecular electrostatic potential and electron density surfaces. These studies show clearly that for the kind of systems presented here, gas phase or implicit sol- vent phase studies are inadequate in explaining realistic situations. Calculations with solvent molecules, even if a few only, are necessary to substantiate experimental observations.