Showing papers in "Computers & Graphics in 2017"

TL;DR: A new, general, and efficient method for unstructured point cloud labeling using fully convolutional networks which is suitable for various types of point clouds such as Lidar or photogrammetric data is described.

TL;DR: The proposed method aims at capturing feature continuity of 3D models, while simultaneously minimizing data preprocessing via the construction of an augmented image representation in an ensemble of convolutional neural networks.

TL;DR: This paper proposes a lightweight volumetric CNN architecture (namely, LightNet) to address the real-time 3D object recognition problem leveraging on multitask learning, and uses LightNet to simultaneously predict class and orientation labels from complete and partial shapes.

TL;DR: A multi-view recurrent neural network (MV-RNN) approach for 3D mesh segmentation that combines the convolutional neural networks and a two-layer long short term memory to yield coherent segmentation of 3D shapes is introduced.

TL;DR: A visual analytics approach is proposed that helps additive manufacturing researchers acquire a deep understanding of patterns in log and imagery data collected by 3D printers to discover patterns related to defects and system performance issues.

TL;DR: This work builds on the well known functional map framework and investigates a novel embedding for the alignment of shapes, which identifies points with their Green’s functions of the Laplace–Beltrami operator, and hence, embed shapes into their own function space.

TL;DR: A comprehensive overview of the digital simulation of hand-made stippling can be found in this article, where a plethora of automatic or semi-automatic algorithms have been proposed over the years.

TL;DR: This paper introduces a novel mobile augmented guide for the Casa Batllo museum (Barcelona, Spain) which is based on a combination of traditional and indirect AR, and further explores the use of indirect AR for the interior of buildings as, at present, this technology is usually used for outdoor environments.

TL;DR: A crosswalk classification system based on deep learning that is robust to the many variations of image and scenarios, presenting a consistent behavior and a comparison study of models trained using fully-automatic data acquisition and annotation against models that were partially annotated.

TL;DR: This report surveys and classifies recent developments in parametric 3D body shape modeling, elucidating the key similarities and differences between existing methods as an aid to understanding their relationships and highlighting opportunities for future research.

TL;DR: A novel simulator for two types of sonars used in underwater applications, using the OpenGL shading language chain on GPU for real-time underwater imaging sonar simulation and generates visually realistic sonar images when compared with real-worldSonar images of the same scenes.

TL;DR: In this paper, the average individual is not good at distinguishing original from edited images, answering correctly on 58% of all images, and only identifying the modified ones 46.5% of the time.

TL;DR: This paper proposes a unified framework that uses the textural content of the images to guide the color transfer and colorization, and introduces an edge-aware texture descriptor based on region covariance, allowing for local color transformations.

TL;DR: This paper presents a new method to model plausible trees with fine details from airborne LiDAR point clouds and shows that it can generate a plausible tree model in less than one second, in addition to preserving features of the original point cloud.

TL;DR: This study shows that tone expansion of stylized content takes the form of gamma correction and a novel EO that adapts the gamma value to the intended style of the video, and proposes a simple color correction method that can be applied after tone expansion to emulate the intended colors in HDR.

TL;DR: This work proposes using a parametric face representation as a shape prior to constrain the estimated 3D face shape, and shows that this process is both faster and provides superior 3D reconstruction accuracy when compared to existing alternatives.

TL;DR: This work surveys the field and presents a taxonomy for classifying existing and new comparison visualization techniques for such data into four fundamental approaches: Juxtaposition, Superimposition, Interchangeable, and Explicit Encoding.

TL;DR: A new taxonomy to classify existing selection techniques is introduced and PRECIOUS is proposed, a novel mid-air technique for selecting out-of-reach objects, featuring iterative refinement in Virtual Reality, an hitherto untried approach in this context.

TL;DR: The results show that participants were able to assess relative distances more precisely and were most confident using the illustrative visualization approach, and a novel and fast real-time hatching scheme that outperforms state-of-the-art methods regarding computation time and memory usage.

TL;DR: A particle-based drag force is proposed to compute airfluid and airrigid interactions to improve plausibility of fluid and rigid behavior and the practicality of the approach is shown by combining it with different types of SPH fluid solvers and by simulating multiple, complex scenes.

TL;DR: One more objective quality metric, UDQM, is added as an additional metric in the evaluation of HDR deghosting algorithms and correlation between objective metric and subjective scores show that HDR-VDP-2 is the most reliable metric for evaluating HDR de Ghosting algorithms.

TL;DR: This work presents visual designs to help understanding a soccer match from the spatiotemporal position of players, and revisits a technique used for flow analysis to help identify where does a player move from a given region in the field.

TL;DR: The results show that, even when a reduced number of finger gestures are used, the synthesized motion can be described as perceptually consistent with the ground truth motion data.

TL;DR: A relief cutting method to turn rigid planar surfaces into flexible ones using meander patterns that consist of interlocked Archimedean spirals is presented and this approach provides a simple interface to construct desired patterns.

TL;DR: A mais recente obra de Wolfgang Streeck, diretor emerito do Instituto Max Planck, oferece uma perspectiva original acerca da globalizacao financeira contemporânea, no qual crises e conflitos serao cada vez mais recorrentes.

TL;DR: This work proposes a set of benchmark images, representing a range of possible subject matter and image features of interest to researchers, and describes the policies, tradeoffs, and reasoning that led to the particular images in the set.

TL;DR: A struts elimination strategy is presented to avoid both inner and outer impending nodes, and angles between adjacent inner struts and radii of struts are further optimized to reduce printing material while achieving a given structural strength.

TL;DR: A non-parametric camera-based calibration for optical see-through glasses is proposed which calibrates the display pixels individually based on the pixels viewing rays instead of triangulating their viewpoint-dependent virtual positions.