Short-Term Realization of the ITRS with GNSS/SLR/VLBI/DORIS

<p>International Terrestrial Reference Frame (ITRF) is the realization of the International Terrestrial Reference System (ITRS), which can be used for the variety applications such as earth research, surveying and mapping. 2000 National Geodetic Coordinate System (CGCS2000) has been established and widely applied as a long-term reference frame in China, however, a software for short-term reference frame establishment is also developed to provide high accuracy applications based on the fusion of GNSS/SLR/VLBI/DORIS. We analyzed the covariance from sinex format of the GNSS/SLR/VLBI/DORIS and the quality of local ties. The errors between the local ties and the ITRF2014 within 1cm was 89% in north direction and 85% east direction. We used inner constraints as the method of datum realization and Helmert variance component estimation for giving the weight of different space geodetic GNSS/SLR/VLBI/DORIS. Finaly, short-term terrestrial reference frame realization software can produce the weekly, monthly and annual frame products for high accuracy applications.</p>

Simultaneous Semantic Segmentation and Depth Completion with Constraint of Boundary

As the core task of scene understanding, semantic segmentation and depth completion play a vital role in lots of applications such as robot navigation, AR/VR and autonomous driving. They are responsible for parsing scenes from the angle of semantics and geometry, respectively. While great progress has been made in both tasks through deep learning technologies, few works have been done on building a joint model by deeply exploring the inner relationship of the above tasks. In this paper, semantic segmentation and depth completion are jointly considered under a multi-task learning framework. By sharing a common encoder part and introducing boundary features as inner constraints in the decoder part, the two tasks can properly share the required information from each other. An extra boundary detection sub-task is responsible for providing the boundary features and constructing cross-task joint loss functions for network training. The entire network is implemented end-to-end and evaluated with both RGB and sparse depth input. Experiments conducted on synthesized and real scene datasets show that our proposed multi-task CNN model can effectively improve the performance of every single task.

Automatic Texture Mapping Algorithm for Plant Leaf Model

Texture mapping of plant leaf is a challenging task because of the shape diversity. Based on the barycentric mapping parameterization method, a novel algorithm is proposed. A suitable boundary mapping is computed by automatically extracting and matching the feature points on boundaries of the mesh and image. The matching process consists of two steps, which utilize the symmetry and curvature information respectively. Through iteratively minimizing a weighted energy, the inner constraints given by users in order to make the details clearer can be satisfied by the presented algorithm. Experiments show that this robust algorithm can produce plant leaves with fine texture details and increase the realism of the models.