ShowMeAI日报系列全新升级！覆盖AI人工智能 工具&框架 | 项目&代码 | 博文&分享 | 数据&资源 | 研究&论文 等方向。点击查看 历史文章列表，在公众号内订阅话题 #ShowMeAI资讯日报，可接收每日最新推送。点击 专题合辑&电子月刊 快速浏览各专题全集。点击 这里 回复关键字 日报 免费获取AI电子月刊与资料包。
1.工具&框架
工具：ONNX YOLOv6目标检测
‘ONNX YOLOv6 Object Detection - Python scripts performing object detection using the YOLOv6 model in ONNX.’ by Ibai Gorordo
GitHub: https://github.com/ibaiGorordo/ONNX-YOLOv6-Object-Detection
工具平台：GitHub搜索引擎
‘GHSearch Platform - GitHub Search Engine: Web Application used to retrieve, store and present projects from GitHub, as well as any statistics related to them.’ by SEART - SoftwarE Analytics Research Team
地址: https://seart-ghs.si.usi.ch/
GitHub: https://github.com/seart-group/ghs
工具：Tooll 3 - 实时动画创建工具包
‘Tooll 3 - an open source software to create realtime motion graphics.’ by Still
GitHub: https://github.com/still-scene/t3
工具库：neuralforecast - 可扩展用户友好的时间序列神经网络预测算法库
‘neuralforecast - Scalable and user friendly neural forecasting algorithms for time series data’ by Nixtla
GitHub: https://github.com/Nixtla/neuralforecast
工具：JupyterLab Desktop - 基于Electron的JupyterLab桌面应用
‘JupyterLab Desktop - JupyterLab desktop application, based on Electron.’ by JupyterLab
GitHub: https://github.com/jupyterlab/jupyterlab-desktop
2.博文&分享
总结：Linux命令参考大全
Every Linux Command I know A-zzzzz - Every Linux Command I know A-Z by 0xTRAW
GitHub: https://github.com/0xTRAW/Linux-Commands-A-Z
3.数据&资源
教程：汇编通俗入门
Some Assembly Required - An approachable introduction to assembly. by Hack Club
GitHub: https://github.com/hackclub/some-assembly-required
4.研究&论文
可以点击 这里 回复关键字 日报，免费获取整理好的论文合辑。
论文：Benchmarking and Analyzing Point Cloud Classification under Corruptions
论文标题：Benchmarking and Analyzing Point Cloud Classification under Corruptions
论文时间：7 Feb 2022
所属领域：点云
对应任务：Classification，Point Cloud Classification，点云分类
论文地址：https://arxiv.org/abs/2202.03377
代码实现：https://github.com/jiawei-ren/modelnetc , https://github.com/ldkong1205/PointCloud-C
论文作者：Jiawei Ren, Liang Pan, Ziwei Liu
论文简介：3D perception, especially point cloud classification, has achieved substantial progress./三维感知，尤其是点云分类，已经取得了实质性的进展。
论文摘要：3D perception, especially point cloud classification, has achieved substantial progress. However, in real-world deployment, point cloud corruptions are inevitable due to the scene complexity, sensor inaccuracy, and processing imprecision. In this work, we aim to rigorously benchmark and analyze point cloud classification under corruptions. To conduct a systematic investigation, we first provide a taxonomy of common 3D corruptions and identify the atomic corruptions. Then, we perform a comprehensive evaluation on a wide range of representative point cloud models to understand their robustness and generalizability. Our benchmark results show that although point cloud classification performance improves over time, the state-of-the-art methods are on the verge of being less robust. Based on the obtained observations, we propose several effective techniques to enhance point cloud classifier robustness. We hope our comprehensive benchmark, in-depth analysis, and proposed techniques could spark future research in robust 3D perception.
3D感知，特别是点云分类，已经取得了实质性的进展。然而，在现实世界的部署中，由于场景的复杂性、传感器的不精确性和处理的不精确性，点云的损坏是不可避免的。在这项工作中，我们的目标是对损坏情况下的点云分类进行严格的基准测试和分析。为了进行系统的调查，我们首先提供了一个常见的三维损坏的分类法，并确定了原子损坏的情况。然后，我们对各种有代表性的点云模型进行了综合评估，以了解其鲁棒性和通用性。我们的基准结果显示，尽管点云分类性能随着时间的推移而提高，但最先进的方法正处于不太稳健的边缘。基于所获得的观察，我们提出了几个有效的技术来提高点云分类器的鲁棒性。我们希望我们全面的基准、深入的分析和提出的技术能够激发未来在鲁棒性三维感知方面的研究。
论文：Data-Driven Denoising of Accelerometer Signals
论文标题：Data-Driven Denoising of Accelerometer Signals
论文时间：13 Jun 2022
所属领域：计算机视觉
对应任务：Denoising，降噪
论文地址：https://arxiv.org/abs/2206.05937
代码实现：https://github.com/ansfl/MEMS-IMU-Denoising
论文作者：Daniel Engelsman, Itzik Klein
论文简介：Modern navigation solutions are largely dependent on the performances of the standalone inertial sensors, especially at times when no external sources are available./现代导航解决方案在很大程度上取决于独立惯性传感器的性能，尤其是在没有外部资源的时候。
论文摘要：Modern navigation solutions are largely dependent on the performances of the standalone inertial sensors, especially at times when no external sources are available. During these outages, the inertial navigation solution is likely to degrade over time due to instrumental noises sources, particularly when using consumer low-cost inertial sensors. Conventionally, model-based estimation algorithms are employed to reduce noise levels and enhance meaningful information, thus improving the navigation solution directly. However, guaranteeing their optimality often proves to be challenging as sensors performance differ in manufacturing quality, process noise modeling, and calibration precision. In the literature, most inertial denoising models are model-based when recently several data-driven approaches were suggested primarily for gyroscope measurements denoising. Data-driven approaches for accelerometer denoising task are more challenging due to the unknown gravity projection on the accelerometer axes. To fill this gap, we propose several learning-based approaches and compare their performances with prominent denoising algorithms, in terms of pure noise removal, followed by stationary coarse alignment procedure. Based on the benchmarking results, obtained in field experiments, we show that: (i) learning-based models perform better than traditional signal processing filtering; (ii) non-parametric kNN algorithm outperforms all state of the art deep learning models examined in this study; (iii) denoising can be fruitful for pure inertial signal reconstruction, but moreover for navigation-related tasks, as both errors are shown to be reduced up to one order of magnitude.
现代导航解决方案在很大程度上取决于独立惯性传感器的性能，尤其是在没有外部资源可用的时候。在这些中断期间，惯性导航解决方案很可能会由于工具性噪声源而随着时间的推移而退化，特别是在使用消费性低成本惯性传感器时。传统上，基于模型的估计算法被用来降低噪音水平和增强有意义的信息，从而直接改善导航解决方案。然而，由于传感器的性能在制造质量、工艺噪声建模和校准精度方面存在差异，保证它们的最优性往往被证明是具有挑战性的。在文献中，大多数惯性去噪模型都是基于模型的，而最近有几个数据驱动的方法被建议主要用于陀螺仪测量去噪。由于加速计轴上的未知重力投影，加速计去噪任务的数据驱动方法更具挑战性。为了填补这一空白，我们提出了几种基于学习的方法，并将它们的性能与著名的去噪算法进行了比较，在纯粹的去噪方面，其次是静止的粗略排列程序。基于在现场实验中获得的基准结果，我们表明。(i) 基于学习的模型比传统的信号处理滤波性能更好；(ii) 非参数kNN算法优于本研究中考察的所有最先进的深度学习模型；(iii) 去噪对于纯粹的惯性信号重建是有成效的，但对于导航相关的任务更是如此，因为两种误差都被证明可以减少到一个数量级。
论文：High-Resolution Virtual Try-On with Misalignment and Occlusion-Handled Conditions
论文标题：High-Resolution Virtual Try-On with Misalignment and Occlusion-Handled Conditions
论文时间：28 Jun 2022
所属领域：计算机视觉
对应任务：穿戴设备
论文地址：https://arxiv.org/abs/2206.14180
代码实现：https://github.com/sangyun884/hr-viton
论文作者：Sangyun Lee, Gyojung Gu, Sunghyun Park, Seunghwan Choi, Jaegul Choo
论文简介：Image-based virtual try-on aims to synthesize an image of a person wearing a given clothing item./基于图像的虚拟试穿旨在合成一个穿戴特定服装的人的图像。
论文摘要：Image-based virtual try-on aims to synthesize an image of a person wearing a given clothing item. To solve the task, the existing methods warp the clothing item to fit the person’s body and generate the segmentation map of the person wearing the item, before fusing the item with the person. However, when the warping and the segmentation generation stages operate individually without information exchange, the misalignment between the warped clothes and the segmentation map occurs, which leads to the artifacts in the final image. The information disconnection also causes excessive warping near the clothing regions occluded by the body parts, so called pixel-squeezing artifacts. To settle the issues, we propose a novel try-on condition generator as a unified module of the two stages (i.e., warping and segmentation generation stages). A newly proposed feature fusion block in the condition generator implements the information exchange, and the condition generator does not create any misalignment or pixel-squeezing artifacts. We also introduce discriminator rejection that filters out the incorrect segmentation map predictions and assures the performance of virtual try-on frameworks. Experiments on a high-resolution dataset demonstrate that our model successfully handles the misalignment and the occlusion, and significantly outperforms the baselines. Code is available at https://github.com/sangyun884/hr-viton
基于图像的虚拟试穿的目的是合成一个穿着给定服装的人的图像。为了解决这个问题，现有的方法在将衣服与人融合之前，先对衣服进行扭曲以适应人的身体，并生成穿着该衣服的人的分割图。然而，当翘曲和分割生成阶段在没有信息交流的情况下单独操作时，翘曲的衣服和分割图之间会发生错位，从而导致最终图像中出现人工痕迹。信息断开还导致在被身体部位遮挡的衣服区域附近发生过度的扭曲，即所谓的像素挤压伪影。为了解决这些问题，我们提出了一种新型的试穿条件生成器，作为两个阶段（即翘曲和分割生成阶段）的统一模块。条件生成器中新提出的特征融合块实现了信息交换，而且条件生成器不会产生任何错位或像素挤压的伪影。我们还引入了判别器处理，过滤掉不正确的分割图预测，保证了虚拟试戴框架的性能。在高分辨率数据集上的实验表明，我们的模型成功地处理了错位和遮挡问题，并大大超过了基线的性能。代码可在 https://github.com/sangyun884/hr-viton
论文：Deepchecks: A Library for Testing and Validating Machine Learning Models and Data
论文标题：Deepchecks: A Library for Testing and Validating Machine Learning Models and Data
论文时间：16 Mar 2022
所属领域：机器学习
论文地址：https://arxiv.org/abs/2203.08491
代码实现：https://github.com/deepchecks/deepchecks
论文作者：Shir Chorev, Philip Tannor, Dan Ben Israel, Noam Bressler, Itay Gabbay, Nir Hutnik, Jonatan Liberman, Matan Perlmutter, Yurii Romanyshyn, Lior Rokach
论文简介：This paper presents Deepchecks, a Python library for comprehensively validating machine learning models and data./本文介绍了Deepchecks，一个用于全面验证机器学习模型和数据的Python库。
论文摘要：This paper presents Deepchecks, a Python library for comprehensively validating machine learning models and data. Our goal is to provide an easy-to-use library comprising of many checks related to various types of issues, such as model predictive performance, data integrity, data distribution mismatches, and more. The package is distributed under the GNU Affero General Public License (AGPL) and relies on core libraries from the scientific Python ecosystem: scikit-learn, PyTorch, NumPy, pandas, and SciPy. Source code, documentation, examples, and an extensive user guide can be found at https://github.com/deepchecks/deepchecks and https://docs.deepchecks.com/
本文介绍了Deepchecks，一个用于全面验证机器学习模型和数据的Python库。我们的目标是提供一个易于使用的库，其中包括许多与各种类型的问题有关的检查，如模型预测性能、数据完整性、数据分布不匹配，等等。该软件包在GNU Affero通用公共许可证（AGPL）下发布，并依赖于科学Python生态系统的核心库：scikit-learn、PyTorch、NumPy、pandas和SciPy。源代码、文档、例子和大量的用户指南可以在 https://github.com/deepchecks/deepchecks 和 https://docs.deepchecks.com/ 上找到。
论文：CONVIQT: Contrastive Video Quality Estimator
论文标题：CONVIQT: Contrastive Video Quality Estimator
论文时间：29 Jun 2022
所属领域：计算机视觉
对应任务：Self-Supervised Learning，Video Quality Assessment，Visual Question Answering，自监督学习，视频质量评估，视觉问答
论文地址：https://arxiv.org/abs/2206.14713
代码实现：https://github.com/pavancm/conviqt
论文作者：Pavan C. Madhusudana, Neil Birkbeck, Yilin Wang, Balu Adsumilli, Alan C. Bovik
论文简介：Perceptual video quality assessment (VQA) is an integral component of many streaming and video sharing platforms./感知性视频质量评估（VQA）是许多流媒体和视频共享平台的一个组成部分。
论文摘要：Perceptual video quality assessment (VQA) is an integral component of many streaming and video sharing platforms. Here we consider the problem of learning perceptually relevant video quality representations in a self-supervised manner. Distortion type identification and degradation level determination is employed as an auxiliary task to train a deep learning model containing a deep Convolutional Neural Network (CNN) that extracts spatial features, as well as a recurrent unit that captures temporal information. The model is trained using a contrastive loss and we therefore refer to this training framework and resulting model as CONtrastive VIdeo Quality EstimaTor (CONVIQT). During testing, the weights of the trained model are frozen, and a linear regressor maps the learned features to quality scores in a no-reference (NR) setting. We conduct comprehensive evaluations of the proposed model on multiple VQA databases by analyzing the correlations between model predictions and ground-truth quality ratings, and achieve competitive performance when compared to state-of-the-art NR-VQA models, even though it is not trained on those databases. Our ablation experiments demonstrate that the learned representations are highly robust and generalize well across synthetic and realistic distortions. Our results indicate that compelling representations with perceptual bearing can be obtained using self-supervised learning. The implementations used in this work have been made available at https://github.com/pavancm/conviqt
感知视频质量评估（VQA）是许多流媒体和视频共享平台的一个组成部分。在这里，我们考虑了以自监督的方式学习感知相关的视频质量表示的问题。失真类型的识别和退化程度的确定被作为一项辅助任务来训练深度学习模型，该模型包含一个提取空间特征的深度卷积神经网络（CNN）以及一个捕捉时间信息的循环单元。该模型使用对比性损失进行训练，因此我们把这个训练框架和产生的模型称为CONtrastive VIdeo Quality EstimaTor（CONVIQT）。在测试过程中，训练好的模型的权重被冻结，一个线性回归器将学到的特征映射到无参考（NR）环境下的质量分数。我们通过分析模型预测和地面真实质量评分之间的相关性，在多个VQA数据库上对所提出的模型进行了全面评估，与最先进的NR-VQA模型相比，取得了具有竞争力的性能，尽管它没有在这些数据库上进行训练。我们的消融实验表明，所学到的表征是高度稳健的，并且在合成和现实的扭曲中具有良好的概括性。我们的结果表明，使用自监督学习可以获得具有感知能力的令人信服的表征。这项工作中使用的实现方法已在 https://github.com/pavancm/conviqt 发布。
论文：ProGen2: Exploring the Boundaries of Protein Language Models
论文标题：ProGen2: Exploring the Boundaries of Protein Language Models
论文时间：27 Jun 2022
所属领域：医学
论文地址：https://arxiv.org/abs/2206.13517
代码实现：https://github.com/salesforce/progen
论文作者：Erik Nijkamp, Jeffrey Ruffolo, Eli N. Weinstein, Nikhil Naik, Ali Madani
论文简介：Attention-based models trained on protein sequences have demonstrated incredible success at classification and generation tasks relevant for artificial intelligence-driven protein design./在蛋白质序列上训练的基于注意力的模型在与人工智能驱动的蛋白质设计有关的分类和生成任务中表现出令人难以置信的成功。
论文摘要：Attention-based models trained on protein sequences have demonstrated incredible success at classification and generation tasks relevant for artificial intelligence-driven protein design. However, we lack a sufficient understanding of how very large-scale models and data play a role in effective protein model development. We introduce a suite of protein language models, named ProGen2, that are scaled up to 6.4B parameters and trained on different sequence datasets drawn from over a billion proteins from genomic, metagenomic, and immune repertoire databases. ProGen2 models show state-of-the-art performance in capturing the distribution of observed evolutionary sequences, generating novel viable sequences, and predicting protein fitness without additional finetuning. As large model sizes and raw numbers of protein sequences continue to become more widely accessible, our results suggest that a growing emphasis needs to be placed on the data distribution provided to a protein sequence model. We release the ProGen2 models and code at https://github.com/salesforce/progen
在蛋白质序列上训练的基于注意力的模型在与人工智能驱动的蛋白质设计有关的分类和生成任务中表现出了令人难以置信的成功。然而，我们对非常大规模的模型和数据如何在有效的蛋白质模型开发中发挥作用缺乏足够的了解。我们介绍了一套蛋白质语言模型，命名为ProGen2，该模型被扩展到64亿个参数，并在不同的序列数据集上进行训练，这些数据集来自基因组、元基因组和免疫剧目数据库的10多亿个蛋白质。ProGen2模型在捕捉观察到的进化序列的分布、生成新的可行序列和预测蛋白质的适应性方面显示出最先进的性能，而不需要额外的微调。随着大的模型规模和蛋白质序列的原始数量继续变得更加广泛，我们的结果表明，需要越来越重视提供给蛋白质序列模型的数据分布。我们将ProGen2模型和代码发布在 https://github.com/salesforce/progen
论文：Denoised MDPs: Learning World Models Better Than the World Itself
论文标题：Denoised MDPs: Learning World Models Better Than the World Itself
论文时间：30 Jun 2022
所属领域：机器学习
对应任务：Representation Learning，表征学习
论文地址：https://arxiv.org/abs/2206.15477
代码实现：https://github.com/facebookresearch/denoised_mdp
论文作者：Tongzhou Wang, Simon S. Du, Antonio Torralba, Phillip Isola, Amy Zhang, Yuandong Tian
论文简介：The ability to separate signal from noise, and reason with clean abstractions, is critical to intelligence./分离信号和噪音的能力，以及用干净的抽象概念进行推理的能力，对智能至关重要。
论文摘要：The ability to separate signal from noise, and reason with clean abstractions, is critical to intelligence. With this ability, humans can efficiently perform real world tasks without considering all possible nuisance factors.How can artificial agents do the same? What kind of information can agents safely discard as noises? In this work, we categorize information out in the wild into four types based on controllability and relation with reward, and formulate useful information as that which is both controllable and reward-relevant. This framework clarifies the kinds information removed by various prior work on representation learning in reinforcement learning (RL), and leads to our proposed approach of learning a Denoised MDP that explicitly factors out certain noise distractors. Extensive experiments on variants of DeepMind Control Suite and RoboDesk demonstrate superior performance of our denoised world model over using raw observations alone, and over prior works, across policy optimization control tasks as well as the non-control task of joint position regression.
将信号与噪音分开，并以干净的抽象概念进行推理的能力，对智能来说至关重要。有了这种能力，人类可以在不考虑所有可能的干扰因素的情况下有效地执行现实世界的任务。人工代理如何做到这一点？什么样的信息可以被代理人安全地作为噪音丢弃？在这项工作中，我们根据可控性和与奖励的关系将野外的信息分为四种类型，并将有用的信息表述为既可控又与奖励相关的信息。这个框架澄清了先前关于强化学习（RL）中的表征学习的各种信息，并导致我们提出了学习去噪MDP的方法，该方法明确地排除了某些噪音干扰因素。在DeepMind控制套件和RoboDesk的变体上进行的广泛实验表明，我们的去噪世界模型在政策优化控制任务以及联合位置回归的非控制任务中，比单独使用原始观测值和先前的工作有更高的性能。
论文：PRANC: Pseudo RAndom Networks for Compacting deep models
论文标题：PRANC: Pseudo RAndom Networks for Compacting deep models
论文时间：16 Jun 2022
论文地址：https://arxiv.org/abs/2206.08464
代码实现：https://github.com/ucdvision/pranc , https://github.com/Guang000/Awesome-Dataset-Distillation
论文作者：Parsa Nooralinejad, Ali Abbasi, Soheil Kolouri, Hamed Pirsiavash
论文简介：PRANC enables 1) efficient communication of models between agents, 2) efficient model storage, and 3) accelerated inference by generating layer-wise weights on the fly./PRANC实现了1）代理之间模型的高效通信，2）高效的模型存储，以及3）通过实时生成层级权重加速推理。
论文摘要：Communication becomes a bottleneck in various distributed Machine Learning settings. Here, we propose a novel training framework that leads to highly efficient communication of models between agents. In short, we train our network to be a linear combination of many pseudo-randomly generated frozen models. For communication, the source agent transmits only the seed scalar used to generate the pseudo-random basis networks along with the learned linear mixture coefficients. Our method, denoted as PRANC, learns almost 100× fewer parameters than a deep model and still performs well on several datasets and architectures. PRANC enables 1) efficient communication of models between agents, 2) efficient model storage, and 3) accelerated inference by generating layer-wise weights on the fly. We test PRANC on CIFAR-10, CIFAR-100, tinyImageNet, and ImageNet-100 with various architectures like AlexNet, LeNet, ResNet18, ResNet20, and ResNet56 and demonstrate a massive reduction in the number of parameters while providing satisfactory performance on these benchmark datasets. The code is available https://github.com/ucdvision/pranc
在各种分布式机器学习环境中，通信成为一个瓶颈。在这里，我们提出了一个新的训练框架，保证了代理之间模型的高效通信。简而言之，我们将网络训练成许多伪随机生成的冻结模型的线性组合。对于通信，源代理只传输用于生成伪随机网络的种子标量以及学到的线性混合系数。我们的方法被称为PRANC，学习的参数几乎比深度模型少100倍，在一些数据集和架构上仍然表现良好。PRANC实现了1）代理之间模型的有效沟通，2）有效的模型存储，以及3）通过在飞行中生成层级权重来加速推理。我们在CIFAR-10、CIFAR-100、tinyImageNet和ImageNet-100上测试了PRANC和各种架构，如AlexNet、LeNet、ResNet18、ResNet20和ResNet56，并证明了参数数量的大量减少，同时在这些基准数据集上提供满意的性能。该代码可在 https://github.com/ucdvision/pranc 获取
我们是 ShowMeAI，致力于传播AI优质内容，分享行业解决方案，用知识加速每一次技术成长！点击查看 历史文章列表，在公众号内订阅话题 #ShowMeAI资讯日报，可接收每日最新推送。点击 专题合辑&电子月刊 快速浏览各专题全集。点击 这里 回复关键字 日报 免费获取AI电子月刊与资料包。