2D Projection Research Articles

On the Simulation of Ultra-Sparse-View and Ultra-Low-Dose Computed Tomography with Maximum a Posteriori Reconstruction Using a Progressive Flow-Based Deep Generative Model.

Ultra-sparse-view computed tomography (CT) algorithms can reduce radiation exposure for patients, but these algorithms lack an explicit cycle consistency loss minimization and an explicit log-likelihood maximization in testing. Here, we propose X2CT-FLOW for the maximum a posteriori (MAP) reconstruction of a three-dimensional (3D) chest CT image from a single or a few two-dimensional (2D) projection images using a progressive flow-based deep generative model, especially for ultra-low-dose protocols. The MAP reconstruction can simultaneously optimize the cycle consistency loss and the log-likelihood. We applied X2CT-FLOW for the reconstruction of 3D chest CT images from biplanar projection images without noise contamination (assuming a standard-dose protocol) and with strong noise contamination (assuming an ultra-low-dose protocol). We simulated an ultra-low-dose protocol. With the standard-dose protocol, our images reconstructed from 2D projected images and 3D ground-truth CT images showed good agreement in terms of structural similarity (SSIM, 0.7675 on average), peak signal-to-noise ratio (PSNR, 25.89 dB on average), mean absolute error (MAE, 0.02364 on average), and normalized root mean square error (NRMSE, 0.05731 on average). Moreover, with the ultra-low-dose protocol, our images reconstructed from 2D projected images and the 3D ground-truth CT images also showed good agreement in terms of SSIM (0.7008 on average), PSNR (23.58 dB on average), MAE (0.02991 on average), and NRMSE (0.07349 on average).

Entropy generation on Casson hybrid nanofluid over a curved stretching sheet with convective boundary condition: Semi-analytical and numerical simulations

The present study reports a computational analysis of entropy generation on the MHD flow behavior of two-dimensional Casson hybrid nanofluid over a porous curved stretching sheet in the appearance of thermal radiation. To examine the rheology of the blood we have incorporated Casson fluid model in which comparison is done for both Newtonian and non-Newtonian models. In this investigation, we used the blood as based and Tantalum ( Ta) and Cobalt ( Co) are nanoparticles due to an extensive range of biomedical applications such as an agent to prevent the heat transfer of blood and wounded tissue, treatments of anemia, and treatment of cancer therapeutics. The governing non-linear coupled partial differential equations are altered into non-linear coupled ordinary differential equations by using the similarity variables. The Homotopy Perturbation Method is used to solve the newly renovated equations semi-analytically. The semi-analytical outcomes are validated with numerical outcomes obtained through the shooting method, and they are also compared to available literature as a limiting case. Graphical projections are supplied with the influence of active parameters for velocity, temperature, Bejan number, entropy production, skin friction, and Nusselt number. The velocity profile increases for higher values of curvature parameter, and mixed convection parameter. The temperature profile increases, when increases the thermal radiation and magnetic parameters. Higher values of the nanoparticle’s volume fraction enhance the rate of heat transfer. Compare to the Newtonian fluid model non-Newtonian provide higher heat transfer.

An Efficient Method of Tooth Segmentation Under Massive Medical Data

To accurately and efficiently complete tooth segmentation from a large amount of oral medical data, the burden of doctors should be reduced. An automatic seed picking method based on 2D projection of the occlusal plane was proposed. First, the authors establish a two-dimensional seed data set for tooth segmentation. Then, this article built a prediction network of teeth seeds based on YOLOv4 to realize the prediction of teeth position as well as the recognition of teeth categories. Finally, according to the statistical optimal seed position, the two-dimensional seeds are calculated and mapped back to the three-dimensional space by the reverse projection transformation method to realize the final picking up of the three-dimensional seeds. Furthermore, combined with the previous work of division line detection, the automatic segmentation of the 3D dental model was realized. The experimental results show that the proposed method has high accuracy and real-time performance, which significantly reduces the burden of human-computer interaction in dental model segmentation.

Deep learning-based dental implant recognition using synthetic X-ray images.

A novel algorithm for generating artificial training samples from triangulated three-dimensional (3D) surface models within the context of dental implant recognition is proposed. The proposed algorithm is based on the calculation of two-dimensional (2D) projections (from a number of different angles) of 3D volumetric representations of computer-aided design (CAD) surface models. A fully convolutional network (FCN) is subsequently trained on the artificially generated X-ray images for the purpose of automatically identifying the connection type associated with a specific dental implant in an actual X-ray image. Semi-automated and fully automated systems are proposed for segmenting questioned dental implants from the background in actual X-ray images. Within the context of the semi-automated system, suitable regions of interest (ROIs), which contain the dental implants, are manually specified. However, as part of the fully automated system, suitable ROIs are automatically detected. It is demonstrated that a segmentation/detection accuracy of 94.0% and a classification/recognition accuracy of 71.7% are attainable within the context of the proposed fully automated system. Since the proposed systems utilise an ensemble of techniques that has not been employed for the purpose of dental implant classification/recognition on any previous occasion, the above-mentioned results are very encouraging.Graphical abstract

Scale issues for geoheritage 3D mapping: The case of Lesvos Geopark, Greece

A geopark can be composed of many individual geosites of various geographical scales, thus, categorization according to cartographic scale is crucial for their 3D mapping. The UNESCO Global Geopark of the island of Lesvos in the north-east Aegean, Greece, is a distinctive example of this type of geopark as it contains many unique geosites that vary in geographical scale. The geographical scale is interconnected with the cartographic scale in which the geosite is visualized. The desired cartographic scale is an essential user requirement within an unmanned aerial vehicle's (UAVs) 3D mapping project as the basis for the data acquisition strategy. This research investigates the scale issues in 3D mapping of geosites. Furthermore, it contributes to the incorporation of the geographic and cartographical scales in association with UAV flight parameters such as Ground Sample Distance (GSD), altitude, gimbal pitch, orientation, and front and side overlapping. A total of 132 geosites located in Lesvos Geopark are being studied to determine the flight parameters of three different UAVs and their camera characteristics. The methodology followed to collect very high-resolution images suitable for 3D mapping consists of five main stages: i) determining the geographical scale of each geosite, ii) defining the cartographic scale of all geosites, iii) calculation of the GSD based on cartographic scale, iv) calculation of UAV flight altitude and flight characteristics, and v) classification of geosites based on the flight characteristics for their 3D mapping. Five geographic (G1: < 0.1 ha, G2: 0.1–1 ha, G3: 1–10 ha, G4: 10–100 ha, G5: >100 ha) and five cartographic (C1: > 1:50, C2: 1:50–1:100, C3: 1:100–1:250, C4: 1:250–1:500, C5: < 1:500) categories were defined based on the geosites' size and extent. The combination of the two scales determines the most efficient flight characteristics and optimally acquires very high-resolution images required for the 3D mapping of the selected geosites. Finally, the categorization and characteristics of flights for data collection for high-resolution 3D mapping are collected and presented in a web application. The web application is addressed to the management board of Lesvos Geopark and supports the decision-making processes on mapping geosites using UAVs.

Design and Analysis of Novel CO2 Conditioning Process in Ship-Based CCS

In this work, CO2 conditioning processes for ship-based CCS sequestration are modelled using the software APSEN HYSYS V11. This study uses the captured CO2 gas from the 3D project as the feed. The feed stream contains water, H2S, and CO as contaminants. The purification processes for dehydration, desulfurization, and CO removal are reviewed. Two liquefaction approaches, the open-cycle and the closed-cycle liquefaction, are modelled and compared for transport pressures 7 and 15 bar. It is found that the energy requirement of the open-cycle process is higher than that of the closed-cycle liquefaction process. For the closed-cycle design, two refrigerants, ammonia and propane, are considered. Results show that the energy requirement of the process using ammonia is lower than that of propane. When comparing the two transport pressures, it is found that liquefaction at 15 bar requires less energy than 7 bar. On top of that, both refrigerants are unsuited for the liquefaction of CO2 at 7 bar, as their operating pressures are below 1 atm. Several optimization concepts are tested on the closed-cycle liquefaction design. The net power consumption of the closed-cycle liquefaction is reduced when CO2 is precooled using the intermediate pressure ammonia streams and the cold from the CO stripper.

Optimal crater landmark selection based on optical navigation performance factors for planetary landing

Planetary craters are natural navigation landmarks that widely exist and are easily observed. Optical navigation based on crater landmarks has become an important autonomous navigation method for planetary landing. Due to the increase in observed crater landmarks and the limitation of onboard computation, the selection of good crater landmarks has gradually become a research hotspot in the field of landmark-based optical navigation. This paper designs a fast crater landmark selection method, which not only considers the configuration observability of crater subsets but also focuses on the influence on navigation performance arising from the measurement uncertainty and the matching confidence of craters, which is different from other landmark selection methods. The factor of measurement uncertainty, which is anisotropic, correlated and nonidentically distributed, is quantified and integrated into selection based on crater pairing detection and localization error evaluation. In addition, the concept of the crater matching confidence factor is introduced, which reflects the possibility of 2D projection measurements corresponding to 3D positions. Combined with the configuration observability factor, the crater landmark selection indicator is formed. Finally, the effectiveness of the proposed method is verified by Monte Carlo simulations.

A NEW CONCEPT OF ZONAL AND SEGMENTAL STRUCTURE OF THE HUMAN KIDNEY

To date, there is no single point of view on the lobar, zonal and segmental structure of the arterial tree of the kidney. There is no generally accepted approach in determining the signs by which the kidney is divided into lobes and/or segments. There is still a discussion regarding the division of the renal artery to daughter branches. The aim of the study was to establish variants of individual anatomical variability of the intra-organ arterial bed of the kidney, taking into account zonal and segmental topography. 116 corrosive preparations of the arterial system of the human kidney were studied. In 3D projections, the number of arterial vessels in the hilum of the kidney, topographic and anatomical features of the trunks of the renal arteries in the hilum of the kidney, variants of division of the renal arteries, zones of local blood supply to areas of the substance of the kidney, depending on the variants of division of the renal artery into branches of the 2nd and 3rd orders. The study was carried out using digital visualization methods, digital image processing and appropriate software. As a result, it was found that the renal artery can be divided into two (two-zonal system of blood supply to the kidney) and three (three-zone system of blood supply to the kidney) zonal arteries. In the case of a bizonal system, the renal artery divides into ventral and dorsal branches (54,2% of cases), as well as superior polar and inferior polar zonal branches (15,5% of cases). In the case of a three-zone system, the following options for the division of the renal artery have been established. The first option - on the upper polar, ventral and dorsal zonal branches (in 13,3% of cases); 2nd option - on the ventral, dorsal and lower polar zonal branches (in 9,4% of cases); 3rd variant - on two ventral and one dorsal zonal branches (in 5,1% of cases) and 4th variant on the superior pole, central and inferior pole zonal arteries (in 2,5% of cases). The results of the study convincingly indicate the need to revise the widespread idea of ​​the five-segment structure of the human kidney.

A Robust Convolutional Neural Network for 6D Object Pose Estimation from RGB Image with Distance Regularization Voting Loss

Six-degree (6D) pose estimation of objects is important for robot manipulation but at the same time challenging when dealing with occluded and textureless objects. To overcome this challenge, the proposed method presents an end-to-end robust network for real-time 6D pose estimation of rigid objects using the RGB image. In this proposed method, a fully convolutional network with a features pyramid is developed that effectively boosts the accuracy of pixelwise labeling and direction unit vector field that take part in the voting process for object keypoints estimation. The network further takes into account measuring the distance between pixel and keypoint, which aims to help select accurate hypotheses in the RANSAC process. This avoids hypothesis deviations caused by the errors due to direction unit vectors in cases of distant pixels from keypoints. A vectorial distance regularization loss function is used to help Perspective-n-Point find 2D-3D correspondences between 3D object keypoints and their estimated corresponding 2D counterparts. Experiments are performed on widely used LINEMOD and occlusion LINEMOD datasets with ADD (-S) and 2D projection evaluation metrics. The results show that our method improves pose estimation performance compared to the state-of-the-art while still achieving real-time efficiency.

PRODUCTION ORGANIZATION OF GRAPHICS PRODUCT

Computer graphics is developing dynamically in various spheres of human life. Paying attention to the high dynamism of the world around us, the issue of optimizing the process of video, animation, character, advertising, avatar, etc. creation is becoming increasingly important. The graphic product requires the fastest production process without paying attention to the fact that it is a development of a feature film, a commercial project of various scales or a project of 10 frames. The analysis of the main achievements of specialists in the field of computer graphics revealed different approaches to the organization of graphic projects. In specialist’s works, attention is paid to the principles of graphic conveyors of major studios and film companies. The purpose of this work is to present recommendations for the organization of the process of developing a short video series by one person, taking into account the automation of the relevant stages of work, on the example of a video of social advertising. To implement this task, the scheme of the project workflow was proposed, where the main stages of creating graphic content are marked, the relevant software and selected parts of the workflow such as preproduction, production and postproduction. The following software tools were selected for the project implementation: Z-Brush, Blender, Autodesk Maya, Adobe After Effects. The analysis of the development algorithm showed the critical points of the scheme, which reduce the speed of the project. To avoid this, it was proposed the technique of "pipeline". As a result of research the methodologies of the graphics pipeline organization was raised, and the beneficial effects of reliable workflow management were illustrated. A visual and detailed form of presentation of the creating a graphic product stages using the "pipeline" technique will help increase the level of knowledge among beginners.

Three-dimensional (3D) optical imaging of electron beam and X-rays from medical linear accelerators (LINAC) using a plastic scintillator plate in water

Although optical imaging of electron beams and X-rays from medical linear accelerators (LINAC) is a possible method for dose distribution measurements, it has been limited to two-dimensional (2D) projection images. For the precise measurement of an optical image of electron beams and X-rays, three-dimensional (3D) imaging is desired. To measure 3D dose distributions, we conducted imaging of electron beams and X-rays using a plastic scintillator plate set in a water phantom. When this plate was immersed in the water phantom, irradiation with electron beams or X-rays was carried out from along the plate's sides. Optical images of the scintillator plate were acquired using a charge-coupled device (CCD) camera from the side during irradiation with electron beams and X-rays. Measurements were conducted at 6 MeV, 9 MeV and 12 MeV for electron beams and at 6 MV and 10 MV for X-rays. The imaging system was set on the bed of the LINAC and moved at 10-mm steps perpendicular to the beam direction to acquire a set of sliced optical images of the beams. A set of these sliced images were stacked and interpolated to form 3D optical images. For the 3D images, after the correction of the Cherenkov-light component in the images, the relative depth and lateral doses were evaluated. From the relative depth doses of electron beams, the half-value depths could be evaluated within an error of 1.3 mm. Lateral widths could be evaluated within an error of less than 2 mm parallel to the plastic scintillator and less than 6.5 mm perpendicular to it. From the relative depth doses of X-rays, the average difference between the measured value and that by a planning system was within an error of 2 %. Lateral widths could be evaluated within an error of less than 0.68 mm parallel to the plastic scintillator and less than 2.6 mm perpendicular to it. We confirmed that 3D imaging of electron beams and X-rays using plastic scintillator plate is feasible and is a promising method for measuring dose images at any position.

Optimizing Full 3D SPARKLING Trajectories for High-Resolution Magnetic Resonance Imaging.

The Spreading Projection Algorithm for Rapid K-space sampLING, or SPARKLING, is an optimization-driven method that has been recently introduced for accelerated 2D MRI using compressed sensing. It has then been extended to address 3D imaging using either stacks of 2D sampling patterns or a local 3D strategy that optimizes a single sampling trajectory at a time. 2D SPARKLING actually performs variable density sampling (VDS) along a prescribed target density while maximizing sampling efficiency and meeting the gradient-based hardware constraints. However, 3D SPARKLING has remained limited in terms of acceleration factors along the third dimension if one wants to preserve a peaky point spread function (PSF) and thus good image quality. In this paper, in order to achieve higher acceleration factors in 3D imaging while preserving image quality, we propose a new efficient algorithm that performs optimization on full 3D SPARKLING. The proposed implementation based on fast multipole methods (FMM) allows us to design sampling patterns with up to 107 k-space samples, thus opening the door to 3D VDS. We compare multi-CPU and GPU implementations and demonstrate that the latter is optimal for 3D imaging in the high-resolution acquisition regime ( 600μ m isotropic). Finally, we show that this novel optimization for full 3D SPARKLING outperforms stacking strategies or 3D twisted projection imaging through retrospective and prospective studies on NIST phantom and in vivo brain scans at 3 Tesla taking the particular case of T2 *-w imaging. Overall the proposed method allows for 2.5-3.75x shorter scan times compared to GRAPPA-4 parallel imaging acquisition at 3 Tesla without compromising image quality.

Geometric feature enhanced line segment extraction from large-scale point clouds with hierarchical topological optimization

As the most common primitives, line segments play an essential role in the vectorized reconstruction of artificial scenes. In this paper, a geometric feature enhanced line extraction method with Hierarchical Topological Optimization is proposed to extract line segments from large-scale point clouds. Firstly, 3D projection plane regions are extracted from point clouds using region growing and merging; Secondly, the point clouds in the detected planar region are projected onto 2D grids for efficiently geometry features enhanced line detection. Specifically, the edge feature index of each point inside each projection grid is calculated and fused to represent the edge-geometry enhanced feature. The likelihood of edge presenting on the projection grids is produced using a pre-trained convolutional neural network that combines multi-scale edge locating outputs. Then, line segments on the projected edge maps are extracted by the MCMLSD algorithm and further back-projected into 3D space to form the 3D line segment candidates. Thirdly, the hierarchical topological relationships between the contour and line segments are used to optimize the candidate 3D line segments. The optimization process consists of merging perceptually accurate 3D line segments, closing the contour line based on the contour and merging plane intersection lines based on line segments. The point clouds of seven large-scale outdoor scenes from the Semantic 3D and WHU-TLS public datasets are selected for qualitative and quantitative evaluations. Experiments show that the proposed approach can efficiently extract line segments that represent the geometric characteristics of the scenes comprehensively and accurately. Compared with the SOTA line segments extraction algorithms (i.e., Lu et al. (2019), Zhang et al. (2020a)), the proposed method filters out major miscellaneous/broken lines and extract more complete line segments. The Completeness and Correctness of the line segments extraction results reach 86% on average compared with manual labeled ground truth (dl, ds: 0.5,0.5), at an average processing speed of 25,000 points per second.

Intelligent Optimization Algorithm of 3D Tracking Technology in Football Player Moving Image Analysis

Computer vision technology began to affect the development of football. There is increasingly high-tech in football broadcast technology, and many application tools have emerged in the field of football broadcast video analysis. The purpose of this paper is to study the improvement of target tracking algorithm for football broadcast video and to study the intelligent optimization algorithm of 3D tracking technology in football player moving image analysis. This paper proposes to select four models of YOLOv5 to perform target detection experiments in football broadcast videos and analyzes the principle of the Deep SORT multitarget tracking algorithm. At the same time, it is based on the 3D tracking and 3D pose estimation of players based on cross-view correlation matching, and to measure the comprehensive performance of the tracker in the football scene, experiments are carried out on the accuracy and speed of the tracker under the football datasets of four different scenes. The experimental results in this paper show that the MOTA values corresponding to the 3D tracking results and 2D projection results obtained in the campus dataset are only 50 and 56.2. This is much lower than the tracking performance when based on other similarity matrices. The MOTA value of the obtained tracking result (92.6) is very close and significantly outperforms other methods. CCOT performs better on datasets 28, 29, 31, and 32, ECO stands out on dataset 38, and Siamese also performs well on datasets 22 and 36.

The Influence of the Fusion State of the Particles during the Simultaneous Impact on an Oxidized Substrate in the Presence of Asperities

The surface areas of the materials are generally the most vulnerable parts since they are directly exposed to different stresses such as wear, fatigue, and corrosion. On the other hand, the successive impact of the droplets finds numerous applications in additive manufacturing technologies such as 3D printing and liquid metal projection. The lamella state formed by the impact and spreading of the first particle represents the target for the second; however, its molten state or fragmented state, bouncing or a slide, interacts thermally and dynamically to allow or not allow stacking. The deposition, deformation, and solidification of the droplets are the constituent steps of the process which determine the final result, as the topography of the substrate not only conditions the mechanical adhesion of the drop to the substrate but also affects the wettability of the surfaces. The objective of this work was to model by numerical simulation the impact of successive and simultaneous droplets with solidification and simultaneous deformation in the presence of an oxide layer on the substrate, which presents different topography, in order to fix the knowledge on the behavior of the impact, the spreading, or the splashing with solidification. However, a peak diameter greater than 5% of the particle diameter causes the formation of concave roughness which directly infects the mechanical adhering and subsequently the formation of an ordinary lamella.

Markerless Point Cloud Matching Algorithm Based on 3D Feature Extraction

Since 3D scanners can only scan in a limited range and the scanning process is prone to occlusion and other problems, a complete 3D model cannot be obtained from a point cloud scanning result. Therefore, point cloud matching is necessary for most 3D scanning projects. We propose a point cloud registration algorithm based on a comprehensive approach, and the research contents include a new curvature-based point cloud feature extraction method, a three-dimensional spatial structure classifier of feature points, and an unmarked point cloud-matching algorithm based on three-dimensional feature extraction. The experiments are based on the simulation data and real data to verify the algorithm and evaluate the accuracy. The experimental results show that the matching accuracy reaches the millimetre level, and the fully automated and high-precision label-free point cloud matching based on 3D features is realized, which can provide innovative and breakthrough help for 3D reconstruction.

Structure Detection in Three-Dimensional Cellular Cryoelectron Tomograms by Reconstructing Two-Dimensional Annotated Tilt Series.

The revolutionary technique cryoelectron tomography (cryo-ET) enables imaging of cellular structure and organization in a near-native environment at submolecular resolution, which is vital to subsequent data analysis and modeling. The conventional structure detection process first reconstructs the three-dimensional (3D) tomogram from a series of two-dimensional (2D) projections and then directly detects subcellular components found within the tomogram. However, this process is challenging due to potential structural information loss during the tomographic reconstruction and the limited scope of existing methods since most major state-of-the-art object detection methods are designed for 2D rather than 3D images. Therefore, in this article, as an alternative approach to complement the conventional process, we propose a novel 2D-to-3D framework that detects structures within 2D projection images before reconstructing the results back to 3D. We implemented the proposed framework as three specific algorithms for three individual tasks: semantic segmentation, edge detection, and object localization. As experimental validation of the 2D-to-3D framework for cryo-ET data, we applied the algorithms to the segmentation of mitochondrial calcium phosphate granules, detection of spherical edges, and localization of mitochondria. Quantitative and qualitative results show better performance for prediction tasks of segmentation on the 2D projections and promising performance on object localization and edge detection, paving the way for future studies in the exploration of cryo-ET for in situ structural biology.

Miniaturized optimal incident light angle-fitted dark field system for contrast-enhanced real-time monitoring of 2D/3D-projected cell motions.

In the field of biology, dark field microscopy provides superior insight into cells and subcellular structures. However, most dark field microscopes are equipped with a dark field filter and a light source on a 2D-based specimen, so only a flat sample can be observed in a limited space. We propose a compact cell monitoring system with built-in dark field filter with an optimized incident angle of the light source to provide real-time cell imaging and spatial cell monitoring for long-term free from phototoxicity. 2D projection imaging was implemented using a modular condenser lens to acquire high-contrast images. This enabled the long-term monitoring of cells, and the real-time monitoring of cell division and death. This system was able to image, by 2D projection, cells on the surface thinly coated with multiwalled carbon nanotubes, as well as living cells that migrated along the surface of glass beads and hydrogel droplets with a diameter of about 160 μm. The optimal incident light angle-fitted dark field system combines high-contrast imaging sensitivity and high spatial resolution to even image cells on 3D surfaces.

An automatic framework for estimating the pose of the catheter distal section using a coarse-to-fine network

Background and objectiveDuring percutaneous coronary intervention procedures, generally only 2D X-ray images are provided. The consequent lack of depth perception makes it difficult for interventionists to visually estimate the pose of medical tools inside the vasculature, especially for novices. Although some automatic methods have been developed to aid interventionists, it is still a challenging task to obtain stable and accurate pose estimation. In this paper, we describe a learning-based framework for estimating the pose of the catheter distal section (CDS). The main innovation of this framework is the proposal of a coarse-to-fine fusion network (CFF-Net) which can achieve the shape and orientation estimation for the CDS. MethodsBy adopting a two-step fusion, CFF-Net progressively solves the shape and orientation ambiguities. The first step is the early fusion where the 2D projection image fuses with the shape prior before input, which makes the estimated result own a specific catheter distal shape. The second step is the late fusion where CFF-Net fuse feature maps and the orientation data from Electromagnetic (EM) sensors to confirm the overall orientation of the CDS. Finally, the estimated pose in the EM space will be obtained after we combine the estimated shape and orientation from CFF-Net with the position information from the EM sensor. ResultsThe effectiveness of CFF-Net has been verified in a simulated environment where RMSE of CFF-Net is 0.706 ± 0.121 mm. This approach was further transferred from simulation to reality using the real-world data, where RMSE of CFF-Net is 1.121 ± 0.124 mm and RMSE of the whole proposed framework is 1.577 ± 0.144 mm. ConclusionIn simulated and real-world experiments, our proposed approach has been proven to achieve high accuracy while ensuring real-time processing for estimating the pose of the CDS.

An Interdisciplinary Design of an Interactive Cultural Heritage Visit for In-Situ, Mixed Reality and Affective Experiences

Interactive technologies, such as mixed-reality and natural interactions with avatars, can enhance cultural heritage and the experience of visiting a museum. In this paper, we present the design rationale of an interactive experience for a cultural heritage place in the church of Roncesvalles at the beginning of Camino de Santiago. We followed a participatory design with a multidisciplinary team which resulted in the design of a spatial augmented reality system that employs 3D projection mapping and a conversational agent acting as the storyteller. Multiple features were identified as desirable for an interactive experience: interdisciplinary design team; in-situ; mixed reality; interactive digital storytelling; avatar; tangible objects; gestures; emotions and groups. The findings from a workshop are presented for guiding other interactive cultural heritage experiences.