National University of Defense Technology

About: National University of Defense Technology is a education organization based out in Changsha, China. It is known for research contribution in the topics: Radar & Synthetic aperture radar. The organization has 39430 authors who have published 40181 publications receiving 358979 citations. The organization is also known as: Guófáng Kēxuéjìshù Dàxué & NUDT.

The TianHe-1A Supercomputer: Its Hardware and Software

Abstract: This paper presents an overview of TianHe-1A (TH-1A) supercomputer, which is built by National University of Defense Technology of China (NUDT). TH-1A adopts a hybrid architecture by integrating CPUs and GPUs, and its interconnect network is a proprietary high-speed communication network. The theoretical peak performance of TH-1A is 4700 TFlops, and its LINPACK test result is 2566 TFlops. It was ranked the No. 1 on the TOP500 List released in November, 2010. TH-1A is now deployed in National Supercomputer Center in Tianjin and provides high performance computing services. TH-1A has played an important role in many applications, such as oil exploration, weather forecast, bio-medical research.

Learning a Recurrent Residual Fusion Network for Multimodal Matching

Abstract: A major challenge in matching between vision and language is that they typically have completely different features and representations. In this work, we introduce a novel bridge between the modality-specific representations by creating a co-embedding space based on a recurrent residual fusion (RRF) block. Specifically, RRF adapts the recurrent mechanism to residual learning, so that it can recursively improve feature embeddings while retaining the shared parameters. Then, a fusion module is used to integrate the intermediate recurrent outputs and generates a more powerful representation. In the matching network, RRF acts as a feature enhancement component to gather visual and textual representations into a more discriminative embedding space where it allows to narrow the crossmodal gap between vision and language. Moreover, we employ a bi-rank loss function to enforce separability of the two modalities in the embedding space. In the experiments, we evaluate the proposed RRF-Net using two multi-modal datasets where it achieves state-of-the-art results.

Learning Canonical Shape Space for Category-Level 6D Object Pose and Size Estimation

Abstract: We present a novel approach to category-level 6D object pose and size estimation. To tackle intra-class shape variations, we learn canonical shape space (CASS), a unified representation for a large variety of instances of a certain object category. In particular, CASS is modeled as the latent space of a deep generative model of canonical 3D shapes with normalized pose. We train a variational auto-encoder (VAE) for generating 3D point clouds in the canonical space from an RGBD image. The VAE is trained in a cross-category fashion, exploiting the publicly available large 3D shape repositories. Since the 3D point cloud is generated in normalized pose (with actual size), the encoder of the VAE learns view-factorized RGBD embedding. It maps an RGBD image in arbitrary view into a poseindependent 3D shape representation. Object pose is then estimated via contrasting it with a pose-dependent feature of the input RGBD extracted with a separate deep neural networks. We integrate the learning of CASS and pose and size estimation into an end-to-end trainable network, achieving the state-of-the-art performance.

Rapid and Near Real-Time Assessments of Population Displacement Using Mobile Phone Data Following Disasters: The 2015 Nepal Earthquake.

Abstract: INTRODUCTION: Sudden impact disasters often result in the displacement of large numbers of people. These movements can occur prior to events, due to early warning messages, or take place post-event due to damages to shelters and livelihoods as well as a result of long-term reconstruction efforts. Displaced populations are especially vulnerable and often in need of support. However, timely and accurate data on the numbers and destinations of displaced populations are extremely challenging to collect across temporal and spatial scales, especially in the aftermath of disasters. Mobile phone call detail records were shown to be a valid data source for estimates of population movements after the 2010 Haiti earthquake, but their potential to provide near real-time ongoing measurements of population displacements immediately after a natural disaster has not been demonstrated. METHODS: A computational architecture and analytical capacity were rapidly deployed within nine days of the Nepal earthquake of 25th April 2015, to provide spatiotemporally detailed estimates of population displacements from call detail records based on movements of 12 million de-identified mobile phones users. RESULTS: Analysis shows the evolution of population mobility patterns after the earthquake and the patterns of return to affected areas, at a high level of detail. Particularly notable is the movement of an estimated 390,000 people above normal from the Kathmandu valley after the earthquake, with most people moving to surrounding areas and the highly-populated areas in the central southern area of Nepal. DISCUSSION: This analysis provides an unprecedented level of information about human movement after a natural disaster, provided within a very short timeframe after the earthquake occurred. The patterns revealed using this method are almost impossible to find through other methods, and are of great interest to humanitarian agencies. Language: en