Depth perception is paramount to tackle real-world problems, ranging from autonomous driving to consumer applications For the latter, depth estimation from a single image represents the most versatile solution, since a standard camera is available on almost any handheld device Nonetheless, two main issues limit its practical deployment: i) the low reliability when deployed in-the-wild and ii) the demanding resource requirements to achieve real-time performance, often not compatible with such devices Therefore, in this paper, we deeply investigate these issues showing how they are both addressable adopting appropriate network design and training strategies -- also outlining how to map the resulting networks on handheld devices to achieve real-time performance Our thorough evaluation highlights the ability of such fast networks to generalize well to new environments, a crucial feature required to tackle the extremely varied contexts faced in real applications Indeed, to further support this evidence, we report experimental results concerning real-time depth-aware augmented reality and image blurring with smartphones in-the-wild

Real-time single image depth perception in the wild with handheld devices

With the development of cloud computing, cloud computing provides more and more computing services. In today's fiercely competitive environment, the service flexibility and choice provided by this highly scalable technology are becoming more and more attractive to enterprises. However, behind the rapid development of cloud computing, its security and trust issues have become increasingly unignorable. The purpose of this paper is to analyze the key technologies of cloud computing security based on the trust model (TM). This paper first proposes a subjective and objective TM based on the ideal point method for the trust problem in the cloud environment, and verifies the performance of the model through experiments. This paper proposes a decentralized TM based on blockchain, and verifies the effectiveness of the model through experiments. This paper starts from the current problems of cloud computing, and on the basis of understanding the shortcomings of traditional security technologies, proposes a key technology of cloud computing security based on the TM. The solution proposed in this paper has a minimum average time consumption of 0.12 ms in the copy + label generation stage, while the minimum average time consumption of a traditional PDP is 16.3 ms. It is obvious that this solution has great advantages.

Shuai Li

Papers

Analysis of Key Technologies of Cloud Computing Security Based on Trust Model