Point clouds are characterized by irregularity and unstructuredness, which pose challenges in efficient data exploitation and discriminative feature extraction. In this paper, we present an unsupervised deep neural architecture called Flattening-Net to represent irregular 3D point clouds of arbitrary geometry and topology as a completely regular 2D point geometry image (PGI) structure, in which coordinates of spatial points are captured in colors of image pixels. \mr{Intuitively, Flattening-Net implicitly approximates a locally smooth 3D-to-2D surface flattening process while effectively preserving neighborhood consistency.} \mr{As a generic representation modality, PGI inherently encodes the intrinsic property of the underlying manifold structure and facilitates surface-style point feature aggregation.} To demonstrate its potential, we construct a unified learning framework directly operating on PGIs to achieve \mr{diverse types of high-level and low-level} downstream applications driven by specific task networks, including classification, segmentation, reconstruction, and upsampling. Extensive experiments demonstrate that our methods perform favorably against the current state-of-the-art competitors. We will make the code and data publicly available at https://github.com/keeganhk/Flattening-Net.
translated by 谷歌翻译
The recent neural implicit representation-based methods have greatly advanced the state of the art for solving the long-standing and challenging problem of reconstructing a discrete surface from a sparse point cloud. These methods generally learn either a binary occupancy or signed/unsigned distance field (SDF/UDF) as surface representation. However, all the existing SDF/UDF-based methods use neural networks to implicitly regress the distance in a purely data-driven manner, thus limiting the accuracy and generalizability to some extent. In contrast, we propose the first geometry-guided method for UDF and its gradient estimation that explicitly formulates the unsigned distance of a query point as the learnable affine averaging of its distances to the tangent planes of neighbouring points. Besides, we model the local geometric structure of the input point clouds by explicitly learning a quadratic polynomial for each point. This not only facilitates upsampling the input sparse point cloud but also naturally induces unoriented normal, which further augments UDF estimation. Finally, to extract triangle meshes from the predicted UDF we propose a customized edge-based marching cube module. We conduct extensive experiments and ablation studies to demonstrate the significant advantages of our method over state-of-the-art methods in terms of reconstruction accuracy, efficiency, and generalizability. The source code is publicly available at https://github.com/rsy6318/GeoUDF.
translated by 谷歌翻译
从嘈杂的点云中恢复高质量的表面,称为点云降级,是几何处理中的一个基本而又具有挑战性的问题。大多数现有方法要么直接将嘈杂的输入或过滤器原始正态变为更新点位置。由点云降解和正常过滤之间的基本相互作用的动机,我们从多任务的角度重新访问点云,并提出一个名为PCDNF的端到端网络,以通过关节正常滤波来denoise点云。特别是,我们引入了一项辅助正常过滤任务,以帮助整体网络更有效地消除噪声,同时更准确地保留几何特征。除了整体体系结构外,我们的网络还具有两个新型模块。一方面,为了提高降噪性能,我们设计了一种形状感知的选择器,以全面考虑学习点,正常特征和几何学先验,以构建特定点的潜在切线空间表示。另一方面,点特征更适合描述几何细节,正常特征更有利于表示几何结构(例如,边缘和角落)。结合点和正常特征使我们能够克服它们的弱点。因此,我们设计一个功能改进模块,以融合点和正常功能,以更好地恢复几何信息。广泛的评估,比较和消融研究表明,所提出的方法在点云降解和正常过滤方面优于最先进的方法。
translated by 谷歌翻译
快速的现场评估(ROSE)技术可以通过适当地分析快速染色的细胞病理学图像来显着加速胰腺癌的诊断。计算机辅助诊断(CAD)可以潜在地解决玫瑰病中病理学家的短缺。但是,不同样品之间的癌性模式差异很大,这使CAD任务极具挑战性。此外,由于不同的染色质量和各种采集装置类型,玫瑰图像在颜色分布,亮度和对比度方面具有复杂的扰动。为了应对这些挑战,我们提出了一种基于随机实例的视觉变压器(SI-VIT)方法,该方法可以减少扰动并增强实例之间的建模。借助重新组装的洗牌实例及其行李级软标签,该方法利用回归头将模型集中在细胞上,而不是各种扰动。同时,该模型与分类头结合在一起,可以有效地识别不同实例之间的一般分布模式。结果表明,分类准确性有了更准确的注意区域的显着提高,表明玫瑰图像的多种模式有效地提取了,并且复杂的扰动大大降低。这也表明SI-VIT在分析细胞病理学图像方面具有巨大的潜力。代码和实验结果可在https://github.com/sagizty/mil-si上获得。
translated by 谷歌翻译
与外部知识的接地对话系统是提高响应质量的一种有希望的方法。大多数现有的作品采用知识图(KGS)作为外部资源,关注对话的最后一句话中实体的贡献,以了解上下文理解和响应。然而,在多转变环境中隐含的知识与公斤关系之间的过渡规律之间的相关性是不足的。为此,我们提出了一个关系过渡意识知识的对话生成模型(RT-KGD)。具体而言,受到人类对话潜在逻辑的启发,我们的模型将对话级别的关系过渡规律与转向级实体语义信息相结合。以这种方式,知识之间的相互作用被认为是产生丰富的线索,以预测适当的知识并产生相干响应。自动评估和手动评估的实验结果表明,我们的模型表现优于最先进的基准。
translated by 谷歌翻译
语音情绪识别(SER)是人类计算机互动的重要组成部分。在本文中,我们提出了一个基于图形同构网络的SER网络,具有加权多个聚合器(WMA-GIN),当邻居节点的特征在杜松子酒结构中聚集在一起时,该网络可以有效地解决信息困惑的问题。此外,采用了一个全粘的(FA)层来减轻过度方面的问题,该问题在包括杜松子酒在内的所有图神经网络(GNN)结构中都存在。此外,采用多相注意机制和多损失训练策略来避免缺少堆叠的WMA轴层中有用的情感信息。我们评估了我们在流行的Iemocap数据集中提出的WMA-GIN的性能。实验结果表明,WMA-GIN的表现优于其他基于GNN的方法,并且可以通过达到72.48%的加权准确性(WA)和67.72%的未加权准确性(UA)来与某些基于高级的基于非冲突的方法相媲美。
translated by 谷歌翻译
大脑网络将大脑区域之间的复杂连接性描述为图形结构,这为研究脑连接素提供了强大的手段。近年来,图形神经网络已成为使用结构化数据的普遍学习范式。但是,由于数据获取的成本相对较高,大多数大脑网络数据集的样本量受到限制,这阻碍了足够的培训中的深度学习模型。受元学习的启发,该论文以有限的培训示例快速学习新概念,研究了在跨数据库中分析脑连接组的数据有效培训策略。具体而言,我们建议在大型样本大小的数据集上进行元训练模型,并将知识转移到小数据集中。此外,我们还探索了两种面向脑网络的设计,包括Atlas转换和自适应任务重新启动。与其他训练前策略相比,我们的基于元学习的方法实现了更高和稳定的性能,这证明了我们提出的解决方案的有效性。该框架还能够以数据驱动的方式获得有关数据集和疾病之间相似之处的新见解。
translated by 谷歌翻译
Mapping the connectome of the human brain using structural or functional connectivity has become one of the most pervasive paradigms for neuroimaging analysis. Recently, Graph Neural Networks (GNNs) motivated from geometric deep learning have attracted broad interest due to their established power for modeling complex networked data. Despite their superior performance in many fields, there has not yet been a systematic study of how to design effective GNNs for brain network analysis. To bridge this gap, we present BrainGB, a benchmark for brain network analysis with GNNs. BrainGB standardizes the process by (1) summarizing brain network construction pipelines for both functional and structural neuroimaging modalities and (2) modularizing the implementation of GNN designs. We conduct extensive experiments on datasets across cohorts and modalities and recommend a set of general recipes for effective GNN designs on brain networks. To support open and reproducible research on GNN-based brain network analysis, we host the BrainGB website at https://braingb.us with models, tutorials, examples, as well as an out-of-box Python package. We hope that this work will provide useful empirical evidence and offer insights for future research in this novel and promising direction.
translated by 谷歌翻译
一方面,在最近的文献中,许多3D多对象跟踪(MOT)的作品集中在跟踪准确性和被忽视的计算速度上,通常是通过设计相当复杂的成本功能和功能提取器来进行的。另一方面,某些方法以跟踪准确性为代价过多地关注计算速度。鉴于这些问题,本文提出了一种强大而快速的基于相机融合的MOT方法,该方法在准确性和速度之间取决于良好的权衡。依靠相机和激光雷达传感器的特性,设计并嵌入了提出的MOT方法中的有效的深层关联机制。该关联机制在对象远处并仅由摄像机检测到2D域中的对象,并在对象出现在LIDAR的视野中以实现平滑融合时获得的2D轨迹进行更新,并更新2D轨迹。 2D和3D轨迹。基于典型数据集的广泛实验表明,就跟踪准确性和处理速度而言,我们提出的方法在最先进的MOT方法上具有明显的优势。我们的代码可公开用于社区的利益。
translated by 谷歌翻译
生成的对抗网络(GANS)已被证明在图像生成任务中非常成功,但GaN培训具有不稳定问题。许多作品通过手动修改GaN架构提高了GaN训练的稳定性,这需要人类专业知识和广泛的试验和错误。因此,目的是自动化模型设计的神经结构搜索(NAS)已经应用于在无条件图像生成的任务上搜索GAN。早期的NAS-GaN仅用于搜索生成器来减少困难。最近的一些作品试图搜索发电机(G)和鉴别器(D)来提高GaN性能,但它们仍然遭受搜索过程中GaN培训的不稳定性。为了缓解不稳定问题,我们提出了一种高效的两阶段进化算法(EA)基于NAS框架来发现GANS,Dubbed \ TextBF {eagan}。具体而言,我们将G和D的搜索分成两个阶段,提出了重量重置策略以提高GaN训练的稳定性。此外,我们执行进展操作以基于多个目标生成帕累托 - 前部架构,导致G和D的优越组合。通过利用重量分享策略和低保真评估,EAGAN可以显着缩短搜索时间。 EAGAN在CIFAR-10上实现了高竞争力的结果(= 8.81 $ \ PM $ 0.10,FID = 9.91),并超越了STL-10数据集上的先前NAS搜索的GAN(= 10.44 $ \ PM $ 0.087,FID = 22.18)。
translated by 谷歌翻译