Search

Articulated skeleton extraction or learning has been extensively studied for 2D (e.g., images and video) and 3D (e.g., volume sequences, motion capture, and mesh sequences) data. Nevertheless, robustly and accurately learning 3D articulated skeletons from point set sequences captured by a single consumer-grade depth camera still remains challenging, since such data are often corrupted with substantial noise and outliers. Relatively few approaches have been proposed to tackle this problem. In this paper, we present a novel unsupervised framework to address this issue. Specifically, we first build one-to-one point correspondences among the point cloud frames in a sequence with our non-rigid point cloud registration algorithm. We then generate a skeleton involving a reasonable number of joints and bones with our skeletal structure extraction algorithm. We lastly present an iterative Linear Blend Skinning based algorithm for accurate joints learning. At the end, our method can learn a quality articulated skeleton from a single 3D point sequence possibly corrupted with noise and outliers. Through qualitative and quantitative evaluations on both publicly available data and in-house Kinect-captured data, we show that our unsupervised approach soundly outperforms state of the art techniques in terms of both quality (i.e., visual) and accuracy (i.e., Euclidean distance error metric). Moreover, the poses of our extracted skeletons are even comparable to those by KinectSDK, a well-known supervised pose estimation technique; for example, our method and KinectSDK achieves similar distance errors of 0.0497 and 0.0521. • We present a novel approach for 3D articulated skeleton extraction using a single depth camera. • We first establish point correspondences among point frames. • We then extract a skeletal structure from the output by the first step. • We finally use the original input and output by the first step for accurate joints learning. • Extensive experiments demonstrate our method outperforms current techniques. [ABSTRACT FROM AUTHOR]

Abstract Transmissible cyber threats have become one of the most serious security issues in cyberspace. Many techniques have been proposed to model, simulate and identify threats' sources and their propagation in large-scale distributed networks. Most techniques are based on the analysis of real networks dataset that contains sensitive information. Traditional in-memory analysis of these dataset often causes data leakage due to system vulnerabilities. If the dataset itself is compromised by adversaries, this threat cost would be even higher than the threat being analysed. In this paper, we propose a new distributed shielded execution framework (Disef) for cyber threats analysis. The Disef framework enables efficient distributed analysis of network dataset while achieves security guarantees of data confidentiality and integrity. In-memory dataset is protected by using a new encrypted key–value format and could be efficiently transferred into Intel SGX enabled enclaves for shielded execution. Our experimental results showed that the proposed framework supports secure in-memory analysis of large network dataset and has comparable performance with systems that have no confidentiality and integrity guarantees. Highlights • A distributed shielded execution framework for network cyber threats analysis is presented. • A brief secure partition mechanism is proposed to decouple secure and normal spaces. • A context-switch secure interface is proposed to reduce Iago attacks. • New key–value encryption operations are integrated to prevent rollback and replay attacks. [ABSTRACT FROM AUTHOR]

Vehicular ad hoc network (VANET) is an increasing important paradigm, which not only provides safety enhancement but also improves roadway system efficiency. However, the security issues of data confidentiality, and access control over transmitted messages in VANET have remained to be solved. In this paper, we propose a secure and efficient message dissemination scheme (SEMD) with policy enforcement in VANET, and construct an outsourcing decryption of ciphertext-policy attribute-based encryption (CP-ABE) to provide differentiated access control services, which makes the vehicles delegate most of the decryption computation to nearest roadside unit (RSU). Performance evaluation demonstrates its efficiency in terms of computational complexity, space complexity, and decryption time. Security proof shows that it is secure against replayable choosen-ciphertext attacks (RCCA) in the standard model. [ABSTRACT FROM AUTHOR]

Existing desktop virtualization systems suffer from a very limited performance in replaying high-definition videos remotely: intolerable CPU and bandwidth consumption, high response delay and poor video quality. In this paper, we propose an original-stream based solution to provide good user experience for replaying high-definition videos in desktop virtualization systems without any modification on applications and support most of prevalent high-definition video formats. In our solution, server׳s video content is not decoded on server but intercepted and delivered to client in its originally encoded state, so that the video content can be easily stored and transported in computer systems with high quality and low bandwidth. The encoded video content is intercepted in server׳s display driver, which enables HDR to work seamlessly with existing applications. The extremely CPU-intensive video decoding tasks are executed on client by using GPU-accelerated video decoding technology so that CPU can concentrate on other tasks. The experimental results validate our method and show that this proposed approach measurably outperforms state-of-the-art solutions. [ABSTRACT FROM AUTHOR]