For the diagnosis of Chinese medicine, tongue segmentation has reached a fairly mature point, but it has little application in the eye diagnosis of Chinese medicine.First, this time we propose Res-UNet based on the architecture of the U2Net network, and use the Data Enhancement Toolkit based on small datasets, Finally, the feature blocks after noise reduction are fused with the high-level features.Finally, the number of network parameters and inference time are used as evaluation indicators to evaluate the model. At the same time, different eye data segmentation frames were compared using Miou, Precision, Recall, F1-Score and FLOPS. To convince people, we cite the UBIVIS. V1 public dataset this time, in which Miou reaches 97.8%, S-measure reaches 97.7%, F1-Score reaches 99.09% and for 320*320 RGB input images, the total parameter volume is 167.83 MB,Due to the excessive number of parameters, we experimented with a small-scale U2Net combined with a Res module with a parameter volume of 4.63 MB, which is similar to U2Net in related indicators, which verifies the effectiveness of our structure.which achieves the best segmentation effect in all the comparison networks and lays a foundation for the application of subsequent visual apparatus recognition symptoms.

Autonomous driving confronts great challenges in complex traffic scenarios, where the risk of Safety of the Intended Functionality (SOTIF) can be triggered by the dynamic operational environment and system insufficiencies. The SOTIF risk is reflected not only intuitively in the collision risk with objects outside the autonomous vehicles (AVs), but also inherently in the performance limitation risk of the implemented algorithms themselves. How to minimize the SOTIF risk for autonomous driving is currently a critical, difficult, and unresolved issue. Therefore, this paper proposes the "Self-Surveillance and Self-Adaption System" as a systematic approach to online minimize the SOTIF risk, which aims to provide a systematic solution for monitoring, quantification, and mitigation of inherent and external risks. The core of this system is the risk monitoring of the implemented artificial intelligence algorithms within the AV. As a demonstration of the Self-Surveillance and Self-Adaption System, the risk monitoring of the perception algorithm, i.e., YOLOv5 is highlighted. Moreover, the inherent perception algorithm risk and external collision risk are jointly quantified via SOTIF entropy, which is then propagated downstream to the decision-making module and mitigated. Finally, several challenging scenarios are demonstrated, and the Hardware-in-the-Loop experiments are conducted to verify the efficiency and effectiveness of the system. The results demonstrate that the Self-Surveillance and Self-Adaption System enables dependable online monitoring, quantification, and mitigation of SOTIF risk in real-time critical traffic environments.

Perception algorithms in autonomous driving systems confront great challenges in long-tail traffic scenarios, where the problems of Safety of the Intended Functionality (SOTIF) could be triggered by the algorithm performance insufficiencies and dynamic operational environment. However, such scenarios are not systematically included in current open-source datasets, and this paper fills the gap accordingly. Based on the analysis and enumeration of trigger conditions, a high-quality diverse dataset is released, including various long-tail traffic scenarios collected from multiple resources. Considering the development of probabilistic object detection (POD), this dataset marks trigger sources that may cause perception SOTIF problems in the scenarios as key objects. In addition, an evaluation protocol is suggested to verify the effectiveness of POD algorithms in identifying the key objects via uncertainty. The dataset never stops expanding, and the first batch of open-source data includes 1126 frames with an average of 2.27 key objects and 2.47 normal objects in each frame. To demonstrate how to use this dataset for SOTIF research, this paper further quantifies the perception SOTIF entropy to confirm whether a scenario is unknown and unsafe for a perception system. The experimental results show that the quantified entropy can effectively and efficiently reflect the failure of the perception algorithm.

最近，通过“向导”模拟游戏收集了一类以任务为导向的对话（TOD）数据集。但是，《巫师》数据实际上是模拟的数据，因此与现实生活中的对话根本不同，这些对话更加嘈杂和随意。最近，Seretod挑战赛是组织的，并发布了Mobilecs数据集，该数据集由来自中国移动的真实用户和客户服务人员之间的真实世界对话框组成。基于Mobilecs数据集，Seretod挑战具有两个任务，不仅评估了对话系统本身的构建，而且还检查了对话框成绩单中的信息提取，这对于建立TOD的知识库至关重要。本文主要介绍了Mobilecs数据集对这两项任务的基线研究。我们介绍了如何构建两个基线，遇到的问题以及结果。我们预计基线可以促进令人兴奋的未来研究，以建立针对现实生活任务的人类机器人对话系统。

Face Animation是计算机视觉中最热门的主题之一，在生成模型的帮助下取得了有希望的性能。但是，由于复杂的运动变形和复杂的面部细节建模，生成保留身份和光真实图像的身份仍然是一个关键的挑战。为了解决这些问题，我们提出了一个面部神经量渲染（FNEVR）网络，以充分探索在统一框架中2D运动翘曲和3D体积渲染的潜力。在FNEVR中，我们设计了一个3D面积渲染（FVR）模块，以增强图像渲染的面部细节。具体而言，我们首先使用精心设计的体系结构提取3D信息，然后引入一个正交自适应射线采样模块以进行有效的渲染。我们还设计了一个轻巧的姿势编辑器，使FNEVR能够以简单而有效的方式编辑面部姿势。广泛的实验表明，我们的FNEVR在广泛使用的说话头基准上获得了最佳的总体质量和性能。

磁共振成像是临床诊断的重要工具。但是，它遭受了漫长的收购时间。深度学习的利用，尤其是深层生成模型，在磁共振成像中提供了积极的加速和更好的重建。然而，学习数据分布作为先验知识并从有限数据中重建图像仍然具有挑战性。在这项工作中，我们提出了一种新颖的Hankel-K空间生成模型（HKGM），该模型可以从一个k-空间数据的训练集中生成样品。在先前的学习阶段，我们首先从k空间数据构建一个大的Hankel矩阵，然后从大型Hankel矩阵中提取多个结构化的K空间贴片，以捕获不同斑块之间的内部分布。从Hankel矩阵中提取斑块使生成模型可以从冗余和低级别的数据空间中学习。在迭代重建阶段，可以观察到所需的解决方案遵守学识渊博的先验知识。通过将其作为生成模型的输入来更新中间重建解决方案。然后，通过对测量数据对其Hankel矩阵和数据一致性组合施加低排名的惩罚来替代地进行操作。实验结果证实，单个K空间数据中斑块的内部统计数据具有足够的信息来学习强大的生成模型并提供最新的重建。

与EMNLP2022 SERETOD车间共同划分的半监督和增强任务的对话系统的挑战。

RNA结构的确定和预测可以促进靶向RNA的药物开发和可用的共性元素设计。但是，由于RNA的固有结构灵活性，所有三种主流结构测定方法（X射线晶体学，NMR和Cryo-EM）在解决RNA结构时会遇到挑战，这导致已解决的RNA结构的稀缺性。计算预测方法作为实验技术的补充。但是，\ textit {de从头}的方法都不基于深度学习，因为可用的结构太少。取而代之的是，他们中的大多数采用了耗时的采样策略，而且它们的性能似乎达到了高原。在这项工作中，我们开发了第一种端到端的深度学习方法E2FOLD-3D，以准确执行\ textit {de de novo} RNA结构预测。提出了几个新的组件来克服数据稀缺性，例如完全不同的端到端管道，二级结构辅助自我鉴定和参数有效的骨干配方。此类设计在独立的，非重叠的RNA拼图测试数据集上进行了验证，并达到平均sub-4 \ aa {}根平方偏差，与最先进的方法相比，它表现出了优越的性能。有趣的是，它在预测RNA复杂结构时也可以取得令人鼓舞的结果，这是先前系统无法完成的壮举。当E2FOLD-3D与实验技术耦合时，RNA结构预测场可以大大提高。

无人机（无人驾驶飞机）动态包围是一个具有巨大潜力的新兴领域。研究人员通常会从生物系统中获得灵感，要么是从宏观世界（如鱼类学校或鸟类羊群）或类似基因调节网络等微世界的灵感。但是，大多数群体控制算法都取决于集中控制，全球信息获取或相邻代理之间的通信。在这项工作中，我们提出了一种纯粹基于视觉的分布式群体控制方法，而没有任何直接通信，例如，群体的代理无人机可以生成一个陷入的模式，以完全基于其安装的全向视觉传感器包围无人机的逃脱目标。还设计了描述每种无人机行为模型的有限状态机器，以便一群无人机可以集体地搜索和捕获目标。我们在各种模拟和现实实验中验证了所提出方法的有效性和效率。

尽管大多数微型机器人在坚固耐用的地形上都面临困难，但甲虫可以在复杂的底物上平稳行走而不会滑倒或粘在地面上，因为它们的刚度可变可变的塔西（Tarsi）和可在塔西（Tarsi）的尖端上伸展的钩子。在这项研究中，我们发现甲虫会积极弯曲并定期扩大爪子以在网状表面上自由爬行。受甲虫的爬行机制的启发，我们设计了一个8厘米的微型攀岩机器人，以与天然甲虫相同的循环方式打开和弯曲的人造爪。机器人可以在网格表面上以可控步态自由攀爬，陡峭的斜角60 {\ deg}，甚至过渡表面。据我们所知，这是第一个可以同时攀登网格表面和悬崖倾斜的微型机器人。