Department of Information Engineering

Time-of-flight (ToF) depth cameras have been increasingly adopted in various real-world applications, e.g., used with RGB cameras for advanced computer vision tasks like 3-D mapping or deployed alone in privacy-sensitive applications such as sleep monitoring. In this paper, we propose UltraDepth, the first system that can expose high-resolution texture from depth maps captured by off-the-shelf ToF cameras, simply by introducing a distorting IR source. The exposed texture information can significantly augment depth-based applications. Moreover, such a capability can be used to launch privacy attacks, which poses a major concern due to the prominence of ToF cameras. To design UltraDepth, we present an in-depth analysis on the impact of the distorting IR light on the distance measurement. We further show that, the reflection properties (reflectivity and incidence angle) of the objects will be encoded in the distorted depth map and hence can be leveraged to reveal texture of objects in UltraDepth. We then propose two practical implementations of UltraDepth, i.e., reflection-based and external IR-based implementations. Our extensive real-world experiments show that, the depth maps output by UltraDepth achieve 89.06%, 99.33%, 81.25% mean accuracy in object detection, face recognition and character recognition, respectively, which offers over 10× improvement over the ordinary depth maps and even approaches the performance of RGB and IR images in a number of scenarios. The findings of this work provide key insights for new research on depth-related computer vision and security of depth sensing devices.

Video popularity ranking is commonly used in various streaming systems. Existing popularity ranking systems usually rank videos from the perspective of access frequency and ignore other implicit feedback, due to the belief that popularity rankings with different implicit metrics are highly consistent. Based on a large collection of real-world system trace data, this paper systematically compares video popularity rankings with four different implicit metrics: access frequency, access user frequency, watching time, normalized watching time. We not only analyze the temporal dynamics of videos' popularities ranking with different metrics, but also examine the correlation between them. We have two main findings. 1) The popularities rankings with access frequency and watching time have strong correlation, but the other pairs have low consistency. 2) The temporal dynamics of daily rankings are similar, while the popularity rankings by access frequency and user number frequency are more stable than others. This paper also proposes a new top-N rank correlation coefficient and proves its efficiency. Our analysis results indicate that the implicit metric of popularity ranking needs to be carefully chosen according to specific applications.

Cloud Storage is a commonly used web application. Its reliability is crucial to personal and SME users. However, we see insufficient cost-effectiveness and reliability of traditional cloud storage method-3 replications. Hence, we proposed multi-temperature storage mechanism, MTSM. It store data, by their accessing frequency, with suitable RAID storage methods. For comparison, we show the calculation and simulation to estimate the MTTDL, utilization rate and cost of MTSM and 3-replicas.

We propose and experimentally investigate a novel WDM-PON architecture using inverse-RZ modulated centralized light sources. The finite optical power in each bit of the downstream IRZ signal can greatly facilitate the upstream data re-modulation.

Variable-bit-rate (VBR) encoded videos can provide a more consistent visual quality than constant-bit-rate (CBR) encoded videos. However, the long-range bit-rate variations in VBR videos make it difficult to efficiently provide quality-of-service control in a video-on-demand system. Existing scheduling algorithms such as Optimal Smoothing, which requires both downward and upward bandwidth reallocations to adapt to the video bit-rate variations, simply cannot guarantee video delivery in networks with mixed video and data traffic. This study tackles this limitation by investigating a new scheduling algorithm with monotonic-decreasing rate allocations for scheduling video data transmissions. By eliminating upward bandwidth reallocations, the proposed scheduler can guarantee video delivery even in the presence of other data traffic in the same network. Moreover, results show that the proposed scheduler can achieve such performance guarantee without tradeoff in performance or resource requirements. This work presents this new monotonic-decreasing rate scheduler, analyzes its fundamental properties, and evaluates its performance using a large number of real-world VBR video traces (274 DVD movies) in extensive trace-driven simulations.