Where is the teacher? Digital analytics for classroom proxemics

The term Classroom Proxemics refers to how teachers and students use classroom space, and the impact of this and the spatial design on learning and teaching. This study addresses the divide between, on the one hand, substantial work on proxemics based on classroom observations and, on the other hand, emerging work to design automated feedback that helps teachers identify salient patterns in their use of the classroom space. This study documents how digital analytics were designed in service of a senior teacher's practice-based inquiry into classroom proxemics. Indoor positioning data from four teachers were analysed, visualized and used as evidence to compare three distinct learning designs enacted in a physics classroom. This study demonstrates how teachers can make effective use of such visualizations, to gain insight into their classroom practice. This is evidenced by (a) documenting teachers' reflections on visualizations of positioning data, both their own and that of peers and (b) identifying the types of indicator (operationalized as analytical metrics) that foreground the most useful information for teachers to gain insight into their practice.     

Learning to detect a salient object

In this paper, we study the salient object detection problem for images. We formulate this problem as a binary labeling task where we separate the salient object from the background. We propose a set of novel features, including multiscale contrast, center-surround histogram, and color spatial distribution, to describe a salient object locally, regionally, and globally. A conditional random field is learned to effectively combine these features for salient object detection. Further, we extend the proposed approach to detect a salient object from sequential images by introducing the dynamic salient features. We collected a large image database containing tens of thousands of carefully labeled images by multiple users and a video segment database, and conducted a set of experiments over them to demonstrate the effectiveness of the proposed approach.     

Highly Scalable Embedded Flash Memory With Deep Trench Isolation and Novel Buried Bitline Integration for the 90-nm Node and Beyond

In this paper, we embedded a Flash memory cell with 90-nm ground-rules in a high-performance CMOS logic process. A novel deep trench isolation (DTI) module enables an isolated p-well (IPW) bias scheme, leading to Flash with uniform channel program/erase by Fowler-Nordheim tunneling without gate induced drain leakage, a key feature for low-power portable electronics. The IPW concept leads to a compact cell design and a highly scalable high-voltage periphery through the narrow intrawell and interwell isolation spaces. The memory arrays are defined by DTI of each bitline (BL) from its neighboring BLs. We additionally present a buried BL (BBL) concept that links the source contacts of each individual BL via the IPW; thus, effectively eliminating one metal line per BL and reducing overall cell size. A conservative cell size shrink of about 40% can be achieved for a uniform channel program/erase-Flash cell with deep trench and BBL compared to a conventional 21F²cell.     

Picture Collage

In this paper, we address a novel problem of automatically creating a picture collage from a group of images. Picture collage is a kind of visual image summary-to arrange all input images on a given canvas, allowing overlay, to maximize visible visual information. We formulate the picture collage creation problem in a conditional random field model, which integrates image salience, canvas constraint, natural preference, and user interaction. Each image is represented by a group of weighted rectangles, which indicate the salient regions. Then picture collage is resolved by minimizing the energy, guided by the constraints. A two-step optimization method is proposed. First, a quick initialization algorithm based on the proposed 1D collage method is presented. Second, a very efficient Markov chain Monte Carlo method is designed for the refined optimization. We also integrate user interaction in the formulation and optimization to obtain an interactive collage reflecting personalized preference. Visual and quantitative experimental evaluations indicate the efficiency of the proposed collage creation technique.     

New CRT sequence sets for a collision channel without feedback

Protocol sequences are binary and periodic sequences used for deterministic multiple access in a collision channel without feedback. In this paper, we focus on user-irrepressible (UI) protocol sequences that can guarantee a positive individual throughput per sequence period with probability one for a slot-synchronous channel, regardless of the delay offsets among the users. As the sequence period has a fundamental impact on the worst-case channel access delay, a common objective of designing UI sequences is to make the sequence period as short as possible. Consider a communication channel that is shared by M active users, and assume that each protocol sequence has a constant Hamming weight w. To attain a better delay performance than previously known UI sequences, this paper presents a CRTm construction of UI sequences with \(w=M+1\), which is a variation of the previously known CRT construction. For all non-prime \(M\ge 8\), our construction produces the shortest known sequence period and the shortest known worst-case delay of UI sequences. Numerical results show that the new construction enjoys a better average delay performance than the optimal random access scheme and other constructions with the same sequence period, in a variety of traffic conditions. In addition, we derive an asymptotic lower bound on the minimum sequence period for \(w=M+1\) if the sequence structure satisfies some technical conditions, called equi-difference, and prove the tightness of this lower bound by using the CRTm construction.

     

Dual-wavelength 10-GHz actively mode-locked erbium fiber laser incorporating highly nonlinear fibers

We demonstrated a dual-wavelength actively mode-locked erbium-doped fiber ring laser with 0.8-nm wavelength spacing at a repetition rate of 10 GHz. A 1-km highly nonlinear fiber is introduced in the nonpolarization-maintaining cavity to suppress the gain competition of the homogeneously broadened gain medium in a self-driven manner. Output pulses at 1557.36 and 1558.17 nm are generated simultaneously. Amplitude fluctuations less than 0.45% and 1.7% of two wavelengths are measured. The corresponding timing jitters are less than 80 and 90 fs, respectively.     

Network externality perspective of feed-in-tariffs (FIT) instruments—Some observations and suggestions

Existing energy policy frameworks revolving around the acceleration of deployment¹ of renewable energy technology can be broadly classified as the quantity vs. price approach. With this brief viewpoint, this paper suggests another perspective of viewing these instruments in terms of a more fundamental basis: whether the deployment in capacity is in terms of a cost minimization approach or a network externality approach. We suggest that the generic price or feed in tariff (FIT) approach in subsidizing renewable electricity generation and associated income would create a bandwagon or self-propagation effect among users rendering the renewable energy technology spreads like a software or information technology. Our objective is to raise awareness of this technology dynamics oriented perspective in renewable deployment supplementing the conventional installation subsidies perspective. We hope that it would inspire more empirical works and studies relating to the policy implications of this viewpoint.     

基于图像的个性化卡通系统

介绍了一个卡通动画系统.该系统可使用户很方便地生成个性化的卡通头像,并通过简单的交互定制各种夸张而生动的表情,还可以实时生成语音驱动的卡通动画.系统主要包括3部分:基于图像的卡通自动生成模块、交互的卡通表情编辑模块和语音驱动的卡通动画模块.卡通自动生成模块采用基于样本学习的方法生成具有特定艺术风格的卡通头像.这里使用了一种非参数化采样算法来学习原始照片和画像间复杂的统计关系.在此基础上,卡通表情编辑器可以通过预先设计的模板生成各种具有生动表情的卡通.同时,用户也可以定制自己的表情模板.卡通动画模块使用了实时的唇形合成算法,可以生成语音驱动的卡通动画.该系统可以使普通用户很方便地生成个性化的卡通头像和动画,在这一点上优于其他动画制作系统.