Yuzhong Shen

This paper presents a real-time interactive external labeling algorithm for automatically placing labels on moderately complex 3D models. The proposed approach considers a set of constraints for the label placement, adjusts the label density on the projected screen, and designs two essential adjustment actions and a two-level label placement strategy based on the sort-and-sweep method to implement the contour-based adaptive external labeling and optimize it as the uniformly distributed label arrangement. This approach operates in real-time by updating the label positions on the fly when the viewport changes. The experimental results show that the proposed method produces high-quality label placement and compares favorably to the state-of-the-art interactive labeling technique, thus providing practical utility for various interactive applications and having great potential for future Virtual Reality (VR) and Augmented Reality (AR) applications.

Deep learning techniques have proven effective in many applications, but these implementations mostly apply to data in one or two dimensions. Handling 3D data is more challenging due to its irregularity and complexity, and there is a growing interest in adapting deep learning techniques to the 3D domain. A recent successful approach called MeshCNN consists of a set of convolutional and pooling operators applied to the edges of triangular meshes. While this approach produced superb results in classification and segmentation of 3D shapes, it can only be applied to edges of a mesh, which can constitute a disadvantage for applications where the focuses are other primitives of the mesh. In this study, we propose face-based and vertex-based operators for mesh convolutional networks. We design two novel architectures based on the MeshCNN network that can operate on faces and vertices of a mesh, respectively. We demonstrate that the proposed face-based architecture outperforms the original MeshCNN implementation in mesh classification and mesh segmentation, setting the new state of the art on benchmark datasets. In addition, we extend the vertex-based operator to fit in the Point2Mesh model for mesh reconstruction from clean, noisy, and incomplete point clouds. While no statistically significant performance improvements are observed, the model training and inference time are reduced by the proposed approach by 91% and 20%, respectively, as compared with the original Point2Mesh model.

Contribution: An approach of integrating financial literacy into an introductory programming course is introduced. New learning materials are developed and the impact of the approach on motivating students and improving learning outcomes are evaluated using mixed methods.

Coastal ecosystems are critically affected by seagrass, both economically and ecologically. However, reliable seagrass distribution information is lacking in nearly all parts of the world because of the excessive costs associated with its assessment. In this paper, we develop two deep learning models for automatic seagrass distribution quantification based on 8-band satellite imagery. Specifically, we implemented a deep capsule network (DCN) and a deep convolutional neural network (CNN) to assess seagrass distribution through regression. The DCN model first determines whether seagrass is presented in the image through classification. Second, if seagrass is presented in the image, it quantifies the seagrass through regression. During training, the regression and classification modules are jointly optimized to achieve end-to-end learning. The CNN model is strictly trained for regression in seagrass and non-seagrass patches. In addition, we propose a transfer learning approach to transfer knowledge in the trained deep models at one location to perform seagrass quantification at a different location. We evaluate the proposed methods in three WorldView-2 satellite images taken from the coastal area in Florida. Experimental results show that the proposed deep DCN and CNN models performed similarly and achieved much better results than a linear regression model and a support vector machine. We also demonstrate that using transfer learning techniques for the quantification of seagrass significantly improved the results as compared to directly applying the deep models to new locations.

A software system has been designed and developed to allow for the display, symbolic manipulation, and player entry of mathematics expressions in mobile games. Display, manipulation, and entry of mathematical expressions are traditionally difficult tasks. Increased limitations on screen space and user input when developing for mobile devices only exacerbate these difficulties. The developed software system balances considerations for ease of use and user interaction with the desire for players to be able to enter answers in a way that is more flexible and interactive than multiple choice, which is the dominant method of interactions in serious games. The software system uses a modular design to separate symbolic math software components from math display components to facilitate reuse of the software system. Additionally, the system displays mathematical expression in a way that is consistent with textbook and handwritten notations to ease the transition between the game and the classroom. Finally, the system provides affordances for natural user interaction to promote usability and engagement. This system has been used to develop a mobile game designed to help students master introductory calculus and physics at the undergraduate level.

Full comprehension of microscopic traffic simulation (MTS) models has necessitated the development of proper visualizations. Existing MTS models only provide limited capability of two- and/or three-dimensional displays that often restrict users’ viewpoint to a flat screen. Their downscaled scenes neither provide a realistic representation of the environment nor allow different users to simultaneously experience the simulation model from different perspectives. This largely prevents analysts from effectively demonstrating and disseminating their simulation results to various stakeholders of different background and knowledge. In light of these issues, this paper aims to develop a framework that enables a multi-user, multi-perspective, and multi-mode ( M³) visualization architecture for microscopic traffic simulation. The proposed framework is empowered by the latest advances in cloud computing and virtual reality (VR) to support interactive and immersive visualization for simulated traffic environments. A client-server architecture allows multiple users at distributed physical locations to view the same simulation from multiple perspectives simultaneously and supports a variety of virtual/augmented reality devices. A prototype of the proposed M³ visualization framework is implemented and demonstrated by simulating and visualizing a model of typical traffic operations in a high-density urban area. The promising capability of the M³ visualization framework is attested. Potential improvements over the present study to further excel current visualization framework are also discussed.

Although Worldview-2 (WV) images (non-pansharpened) have 2-m resolution, the re-visit times for the same areas may be seven days or more. In contrast, Planet images are collected using small satellites that can cover the whole Earth almost daily. However, the resolution of Planet images is 3.125 m. It would be ideal to fuse these two satellites images to generate high spatial resolution (2 m) and high temporal resolution (1 or 2 days) images for applications such as damage assessment, border monitoring, etc. that require quick decisions. In this paper, we evaluate three approaches to fusing Worldview (WV) and Planet images. These approaches are known as Spatial and Temporal Adaptive Reflectance Fusion Model (STARFM), Flexible Spatiotemporal Data Fusion (FSDAF), and Hybrid Color Mapping (HCM), which have been applied to the fusion of MODIS and Landsat images in recent years. Experimental results using actual Planet and Worldview images demonstrated that the three aforementioned approaches have comparable performance and can all generate high quality prediction images.

This paper proposes a novel rigorous transformation model for 2D-3D registration to address the difficult problem of obtaining a sufficient number of well-distributed ground control points (GCPs) in urban areas with tall buildings. The proposed model applies two types of geometric constraints, co-planarity and perpendicularity, to the conventional photogrammetric collinearity model. Both types of geometric information are directly obtained from geometric building structures, with which the geometric constraints are automatically created and combined into the conventional transformation model. A test field located in downtown Denver, Colorado, is used to evaluate the accuracy and reliability of the proposed method. The comparison analysis of the accuracy achieved by the proposed method and the conventional method is conducted. Experimental results demonstrated that: (1) the theoretical accuracy of the solved registration parameters can reach 0.47 pixels, whereas the other methods reach only 1.23 and 1.09 pixels; (2) the RMS values of 2D-3D registration achieved by the proposed model are only two pixels along the x and y directions, much smaller than the RMS values of the conventional model, which are approximately 10 pixels along the x and y directions. These results demonstrate that the proposed method is able to significantly improve the accuracy of 2D-3D registration with much fewer GCPs in urban areas with tall buildings.