Shape transformer nets: Generating viewpoint-invariant 3D shapes from a single image

Single-view 3D shapes generation has achieved great success in recent years. However, current methods always blind the learning of shapes and viewpoints. The generated shape only fit the observed viewpoints and would not be optimal from unknown viewpoints. In this paper, we propose a novel encoder–decoder based network which contains a disentangled transformer to generate the viewpoint-invariant 3D shapes. The differentiable and parametric Non-uniform B-spline (NURBS) surface generation and 3D-to-3D viewpoint transformation are incorporated to learn the viewpoint-invariant shape and the camera viewpoint, respectively. Our new framework allows us to learn the latent geometric parameters of shapes and viewpoints without knowing the ground truth viewpoint. That can simultaneously generate camera-viewpoint and viewpoint-invariant 3D shapes of the object. We analyze the effects of disentanglement and show both quantitative and qualitative results of shapes generated at various unknown viewpoints.     

How to predict the global instantaneous feeling induced by a facial picture?

Picture selection is a time-consuming task for humans and a real challenge for machines, which have to retrieve complex and subjective information from image pixels. An automated system that infers human feelings from digital portraits would be of great help for profile picture selection, photo album creation or photo editing. In this work, two models of facial pictures evaluation are defined. The first one predicts the overall aesthetic quality of a facial image, and the second one answers the question “Among a set of facial pictures of a given person, on which picture does the person look like the most friendly?”. Aesthetic quality is evaluated by the computation of 15 features that encode low-level statistics in different image regions (face, eyes, and mouth). Relevant features are automatically selected by a feature ranking technique, and the outputs of 4 learning algorithms are fused in order to make a robust and accurate prediction of the image quality. Results are compared with recent works and the proposed algorithm obtains the best performance. The same pipeline is considered to evaluate the likability of a facial picture, with the difference that the estimation is based on high-level attributes such as gender, age, and smile. Performance of these attributes is compared with previous techniques that mostly rely on facial keypoint positions, and it is shown that it is possible to obtain likability predictions that are close to human perception. Finally, a combination of both models that selects a likable facial image of good quality for a given person is described.     

以“动态”观点分析判断排除设备故障

主要介绍了“动态”观点分析判断设备故障的简单方法,并且通过2个现场的维修实例来加以说明,如何引导维修人员以“动态”观点分析、判断和排除设备故障,将设备故障以最短的时间内排除,值得借鉴。     

武汉城市总体规划前期研究的实践与思考

新一轮的城市总体规划在时代背景、编制条件和规划统筹等方面发生了深刻变化,坚持科学发展观,提倡政府组织、专家领衔的研究过程,既是对总规技术统筹工作的考验,又是对总规修编组织运作的挑战.结合武汉新一轮城市总体规划的编制实践,介绍了前期研究的组织过程和得失经验,期望能够为大家提供一些有价值的参考.     

Evaluation of the quality scoring system of hard red spring wheat using four different roller mills

Wheat quality laboratories need a scoring system for efficient segregation of wheat samples. A comprehensive and objective scoring system was developed to assist the evaluation of end-use quality of hard red spring (HRS) wheat. The overall quality score consisted of (1) wheat, (2) milling, (3) flour and dough, and (4) baking quality scores. Subscores attained from various quality tests were subdivided into individual categories. The objective of this experiment was to evaluate the quality scoring system. Since flour quality is highly influenced by experimental milling procedures, four different laboratory roller mills were compared for the scoring system using 12 HRS wheat samples. Although the mill type varied significantly (p < 0.001) for the overall quality scores, they appeared to be consistent for ranking of wheat samples, showing significant correlation coefficients (r ≥ 0.74, p < 0.01) between different mill types. Quad. Sr. mill was specifically identified to be more suitable for dough and bread-making quality evaluation of HRS wheat due to the higher consistency with a long-flow MIAG mill. The overall quality scores also showed a significant (p < 0.001) variation among wheat samples, indicating that the scoring system could be useful to segregate HRS wheat samples for end users and breeders.

Abbreviation: HRS: Hard Red Spring; MN: Minnesota; MT: Montana; ND: North Dakota; SD: South Dakota; OVA: Overseas Varietal Analysis; Quad: Quadrumat; AACCI: American Association of Cereal Chemist International; SKB: Sandstedt Kneen and Blish; SAS: Statistical Analysis System; ANOVA: Analysis of Variance; LSD: Least Significant Difference; ERE: Extensograph Resistance to Extension; SDS: Sodium Dodecyl Sulfate     

对欧洲城市与建筑的认知与研究

欧洲九国考察对开阔视野、拓展设计思维有现实意义。本文除国外考察的印象外,还就建筑与环境,欧式建筑形式、风格,雕塑绘画艺术,城市乡村概念,新现代居住建筑视点等提出了五点认知。     

经阴道超声弹性成像鉴别诊断宫颈占位性病变良恶性的价值

目的：探讨经阴道超声弹性成像鉴别诊断宫颈占位性病变良恶性的价值。方法对95例拟诊为宫颈占位性病变患者行经阴道超声弹性成像检查,参照改良5分法评分标准进行评分,并对病变部位进行弹性应变率比值（elastic strain ratios,SRs）检测,将2种方法得到的结果与病理结果对照;SRs 法的 ROC 曲线下面积和最佳临界点通过 ROC 曲线评价。结果95例患者中：病理证实为良性病变45例,SRs 为1．87±0．70;恶性病变50例,SRs 为3．72±0．61,二者 SRs 比较差异有统计学意义（P ＜0．05）。改良5分评分法评分1～2分44例,3～5分51例;SRs法 ROC 曲线下面积为0．982,约登指数最大值为0．896,ROC 曲线最佳临界点为2．83,SRs≥2．8349例,SRs＜2．8346例;SRs 法诊断的敏感性、特异性和准确性分别为94．0％、95．6％和94．7％,改良5分法评分诊断的敏感性、特异性和准确性分别为94．0％、91．1％和92．6％,2种方法对宫颈占位性病变的诊断价值比较,差异无统计学意义（P ＞0．05）。结论经阴道超声弹性成像对宫颈占位性病变良恶性有较高的鉴别诊断价值。     

基于动态标签的关系抽取方法

针对远程监督数据集的关系抽取研究方法存在着大量标签噪声的问题,提出了一种作用于分层注意力机制关系抽取模型的动态标签方法。首先，提出了一种根据关系类别相似性生成动态标签的概念。由于相同的关系标签包含相似的特征信息，计算特征信息的关系类别相似性有助于生成与特征信息相对应的动态标签。其次，利用动态标签方法的评分函数来评价远程监督标签是否为噪声，以决定是否需要生成新的标签代替远程监督标签，通过调整远程监督标签来抑制标签噪声对模型的影响。最后，根据动态标签来更新分层注意力机制以关注有效实例，重新学习每个有效实例的重要性，进一步抽取关键的关系特征信息。实验结果表明，相较于原始的分层注意力机制关系抽取模型，所提方法在Micro和Macro分数上分别有1.3个百分点和1.9个百分点的提升，实现了噪声标签的动态纠正，提升了模型的关系抽取能力。