数据插补是处理缺失数据的有效方法,这在实际应用中很常见。在这项研究中,我们提出并测试一个实现两个重要目标的新型数据归合过程:(1)保留观测值之间的行相似性和功能矩阵中特征之间的列背景关系,以及(2)量身定制插补。处理特定下游标签预测任务。所提出的插补过程使用变压器网络和图形结构学习来迭代地完善观察值之间特征和相似性之间的上下文关系。此外,它使用一个元学习框架来选择对下游预测任务影响的功能。我们对现实世界中的大数据集进行实验,并表明所提出的插补过程始终在各种基准方法上改善插补和标签预测性能。
translated by 谷歌翻译
一阶段算法已被广泛用于需要使用大量数据训练的目标检测系统。他们中的大多数都以实时和准确的态度表现出色。但是,由于其卷积结构,他们需要更多的计算能力和更多的记忆消耗。因此,我们将修剪策略应用于靶向检测网络,以减少参数的数量和模型的大小。为了证明修剪方法的实用性,我们为实验选择了Yolov5模型,并提供了室外障碍的数据集以显示模型的效果。在最佳情况下,在此特定数据集中,与原始模型相比,网络模型的体积减少了49.7%,并且推理时间减少了52.5%。同时,它还使用数据处理方法来补偿修剪导致的准确性下降。
translated by 谷歌翻译
键入的需要图试图从文本中学习谓词之间的零件关系,并将其建模为谓词节点之间的边缘。构造图形通常遭受严重的稀疏性和分布相似性的不可靠性。我们提出了一个两阶段的方法,即带有文本含义和传递性的元素图(EGT2)。EGT2通过识别由键入CCG比较谓词形成的模板句子之间可能的文本构成来学习本地的关系关系。基于生成的本地图,EGT2然后使用三个新型的软传递性约束来考虑构成结构中的逻辑传递性。基准数据集上的实验表明,EGT2可以很好地对需要图中的传递性进行模拟以减轻稀疏性问题,并导致对当前最新方法的显着改善。
translated by 谷歌翻译
本文介绍了使用变压器解决关键点检测和实例关联的新方法。对于自下而上的多人姿势估计模型,他们需要检测关键点并在关键点之间学习关联信息。我们认为这些问题可以完全由变压器解决。具体而言,变压器中的自我关注测量任何一对位置之间的依赖性,这可以为关键点分组提供关联信息。但是,天真的注意力模式仍然没有主观控制,因此无法保证关键点始终会参加它们所属的实例。为了解决它,我们提出了一种监督多人关键点检测和实例关联的自我关注的新方法。通过使用实例掩码来监督自我关注的实例感知,我们可以基于成对引人注定分数为其对应的实例分配检测到的关键字,而无需使用预定义的偏移量字段或嵌入像基于CNN的自下而上模型。我们方法的另一个好处是可以从监督的注意矩阵直接获得任何数量的人的实例分段结果,从而简化了像素分配管道。对Coco多人关键点检测挑战和人实例分割任务的实验证明了所提出的方法的有效性和简单性,并显示出于针对特定目的控制自我关注行为的有希望的方法。
translated by 谷歌翻译
最近的智能故障诊断(IFD)的进展大大依赖于深度代表学习和大量标记数据。然而,机器通常以各种工作条件操作,或者目标任务具有不同的分布,其中包含用于训练的收集数据(域移位问题)。此外,目标域中的新收集的测试数据通常是未标记的,导致基于无监督的深度转移学习(基于UDTL为基础的)IFD问题。虽然它已经实现了巨大的发展,但标准和开放的源代码框架以及基于UDTL的IFD的比较研究尚未建立。在本文中,我们根据不同的任务,构建新的分类系统并对基于UDTL的IFD进行全面审查。对一些典型方法和数据集的比较分析显示了基于UDTL的IFD中的一些开放和基本问题,这很少研究,包括特征,骨干,负转移,物理前导等的可转移性,强调UDTL的重要性和再现性 - 基于IFD,整个测试框架将发布给研究界以促进未来的研究。总之,发布的框架和比较研究可以作为扩展界面和基本结果,以便对基于UDTL的IFD进行新的研究。代码框架可用于\ url {https:/github.com/zhaozhibin/udtl}。
translated by 谷歌翻译
Previous studies have explored generating accurately lip-synced talking faces for arbitrary targets given audio conditions. However, most of them deform or generate the whole facial area, leading to non-realistic results. In this work, we delve into the formulation of altering only the mouth shapes of the target person. This requires masking a large percentage of the original image and seamlessly inpainting it with the aid of audio and reference frames. To this end, we propose the Audio-Visual Context-Aware Transformer (AV-CAT) framework, which produces accurate lip-sync with photo-realistic quality by predicting the masked mouth shapes. Our key insight is to exploit desired contextual information provided in audio and visual modalities thoroughly with delicately designed Transformers. Specifically, we propose a convolution-Transformer hybrid backbone and design an attention-based fusion strategy for filling the masked parts. It uniformly attends to the textural information on the unmasked regions and the reference frame. Then the semantic audio information is involved in enhancing the self-attention computation. Additionally, a refinement network with audio injection improves both image and lip-sync quality. Extensive experiments validate that our model can generate high-fidelity lip-synced results for arbitrary subjects.
translated by 谷歌翻译
Graph anomaly detection (GAD) is a vital task in graph-based machine learning and has been widely applied in many real-world applications. The primary goal of GAD is to capture anomalous nodes from graph datasets, which evidently deviate from the majority of nodes. Recent methods have paid attention to various scales of contrastive strategies for GAD, i.e., node-subgraph and node-node contrasts. However, they neglect the subgraph-subgraph comparison information which the normal and abnormal subgraph pairs behave differently in terms of embeddings and structures in GAD, resulting in sub-optimal task performance. In this paper, we fulfill the above idea in the proposed multi-view multi-scale contrastive learning framework with subgraph-subgraph contrast for the first practice. To be specific, we regard the original input graph as the first view and generate the second view by graph augmentation with edge modifications. With the guidance of maximizing the similarity of the subgraph pairs, the proposed subgraph-subgraph contrast contributes to more robust subgraph embeddings despite of the structure variation. Moreover, the introduced subgraph-subgraph contrast cooperates well with the widely-adopted node-subgraph and node-node contrastive counterparts for mutual GAD performance promotions. Besides, we also conduct sufficient experiments to investigate the impact of different graph augmentation approaches on detection performance. The comprehensive experimental results well demonstrate the superiority of our method compared with the state-of-the-art approaches and the effectiveness of the multi-view subgraph pair contrastive strategy for the GAD task.
translated by 谷歌翻译
We present an optimization-based design process that can generate the best insertion-based joints with respect to different errors, including manipulation error, manufacturing error, and sensing error. We separate the analysis into two stages, the insertion and the after-insertion stability. Each sub-process is discretized into different modes of contacts. The transitions among the contact modes form a directed graph and the connectivity of the graph is achieved and maintained through the manipulation of the socket edge-angle and peg contact-point locations. The analysis starts in 2D with the assumption of point-edge contacts. During the optimization, the edges of the socket are rotated and the points on the peg are moved along the edges to ensure the successful insertion and the stability after insertion. We show in simulation that our proposed method can generate insertion-based joints that are tolerant to the given errors. and we present a few simple 3D projections to show that the analysis is still effective beyond 2D cases.
translated by 谷歌翻译
鉴于其广泛的应用,已经对人面部交换的任务进行了许多尝试。尽管现有的方法主要依赖于乏味的网络和损失设计,但它们仍然在源和目标面之间的信息平衡中挣扎,并倾向于产生可见的人工制品。在这项工作中,我们引入了一个名为StylesWap的简洁有效的框架。我们的核心想法是利用基于样式的生成器来增强高保真性和稳健的面部交换,因此可以采用发电机的优势来优化身份相似性。我们仅通过最小的修改来确定,StyleGAN2体系结构可以成功地处理来自源和目标的所需信息。此外,受到TORGB层的启发,进一步设计了交换驱动的面具分支以改善信息的融合。此外,可以采用stylegan倒置的优势。特别是,提出了交换引导的ID反转策略来优化身份相似性。广泛的实验验证了我们的框架会产生高质量的面部交换结果,从而超过了最先进的方法,既有定性和定量。
translated by 谷歌翻译
受数字孪生系统的启发,开发了一个新型的实时数字双框架,以增强机器人对地形条件的感知。基于相同的物理模型和运动控制,这项工作利用了与真实机器人同步的模拟数字双重同步,以捕获和提取两个系统之间的差异信息,这两个系统提供了多个物理数量的高维线索,以表示代表差异建模和现实世界。柔软的,非刚性的地形会导致腿部运动中常见的失败,因此,视觉感知完全不足以估计地形的这种物理特性。我们使用了数字双重来开发可折叠性的估计,这通过动态步行过程中的物理互动来解决此问题。真实机器人及其数字双重双重测量之间的感觉测量的差异用作用于地形可折叠性分析的基于学习的算法的输入。尽管仅在模拟中受过培训,但学习的模型可以在模拟和现实世界中成功执行可折叠性估计。我们对结果的评估表明,对不同方案和数字双重的优势的概括,可在地面条件下可靠地检测到细微差别。
translated by 谷歌翻译