通常使用卷积神经网络（CNN）进行计算机视觉。 CNN是计算密集型的，并且在移动和互联网（IoT）设备等电力控制系统上部署。 CNN是计算密集型的，因为它们不加选择地计算输入图像的所有像素上的许多特征。我们观察到，鉴于计算机视觉任务，图像通常包含与任务无关的像素。例如，如果任务正在寻找汽车，那么天空中的像素不是很有用。因此，我们建议对CNN进行修改以仅在相关像素上操作以节省计算和能量。我们提出了一种研究三个流行的计算机视觉数据集的方法，发现48％的像素无关紧要。我们还提出了重点卷积，以修改CNN的卷积层，以拒绝明显无关的像素。在嵌入式设备上，我们没有观察到准确性的损失，而推论潜伏期，能耗和倍增add计数均减少了约45％。

Deep neural networks (DNNs) are currently widely used for many artificial intelligence (AI) applications including computer vision, speech recognition, and robotics. While DNNs deliver state-of-the-art accuracy on many AI tasks, it comes at the cost of high computational complexity. Accordingly, techniques that enable efficient processing of DNNs to improve energy efficiency and throughput without sacrificing application accuracy or increasing hardware cost are critical to the wide deployment of DNNs in AI systems.This article aims to provide a comprehensive tutorial and survey about the recent advances towards the goal of enabling efficient processing of DNNs. Specifically, it will provide an overview of DNNs, discuss various hardware platforms and architectures that support DNNs, and highlight key trends in reducing the computation cost of DNNs either solely via hardware design changes or via joint hardware design and DNN algorithm changes. It will also summarize various development resources that enable researchers and practitioners to quickly get started in this field, and highlight important benchmarking metrics and design considerations that should be used for evaluating the rapidly growing number of DNN hardware designs, optionally including algorithmic co-designs, being proposed in academia and industry.The reader will take away the following concepts from this article: understand the key design considerations for DNNs; be able to evaluate different DNN hardware implementations with benchmarks and comparison metrics; understand the trade-offs between various hardware architectures and platforms; be able to evaluate the utility of various DNN design techniques for efficient processing; and understand recent implementation trends and opportunities.

Semantic segmentation works on the computer vision algorithm for assigning each pixel of an image into a class. The task of semantic segmentation should be performed with both accuracy and efficiency. Most of the existing deep FCNs yield to heavy computations and these networks are very power hungry, unsuitable for real-time applications on portable devices. This project analyzes current semantic segmentation models to explore the feasibility of applying these models for emergency response during catastrophic events. We compare the performance of real-time semantic segmentation models with non-real-time counterparts constrained by aerial images under oppositional settings. Furthermore, we train several models on the Flood-Net dataset, containing UAV images captured after Hurricane Harvey, and benchmark their execution on special classes such as flooded buildings vs. non-flooded buildings or flooded roads vs. non-flooded roads. In this project, we developed a real-time UNet based model and deployed that network on Jetson AGX Xavier module.

为了实现不断增长的准确性，通常会开发大型和复杂的神经网络。这样的模型需要高度的计算资源，因此不能在边缘设备上部署。由于它们在几个应用领域的有用性，建立资源有效的通用网络非常感兴趣。在这项工作中，我们努力有效地结合了CNN和变压器模型的优势，并提出了一种新的有效混合体系结构。特别是在EDGENEXT中，我们引入了分裂深度转置注意力（SDTA）编码器，该编码器将输入张量分解为多个通道组，并利用深度旋转以及跨通道维度的自我注意力，以隐含地增加接受场并编码多尺度特征。我们在分类，检测和分割任务上进行的广泛实验揭示了所提出的方法的优点，优于相对较低的计算要求的最先进方法。我们具有130万参数的EDGENEXT模型在Imagenet-1k上达到71.2 \％TOP-1的精度，超过移动设备的绝对增益为2.2 \％，而拖鞋减少了28 \％。此外，我们具有560万参数的EDGENEXT模型在Imagenet-1k上达到了79.4 \％TOP-1的精度。代码和模型可在https://t.ly/_vu9上公开获得。

Despite its importance for federated learning, continuous learning and many other applications, on-device training remains an open problem for EdgeAI. The problem stems from the large number of operations (e.g., floating point multiplications and additions) and memory consumption required during training by the back-propagation algorithm. Consequently, in this paper, we propose a new gradient filtering approach which enables on-device DNN model training. More precisely, our approach creates a special structure with fewer unique elements in the gradient map, thus significantly reducing the computational complexity and memory consumption of back propagation during training. Extensive experiments on image classification and semantic segmentation with multiple DNN models (e.g., MobileNet, DeepLabV3, UPerNet) and devices (e.g., Raspberry Pi and Jetson Nano) demonstrate the effectiveness and wide applicability of our approach. For example, compared to SOTA, we achieve up to 19$\times$ speedup and 77.1% memory savings on ImageNet classification with only 0.1% accuracy loss. Finally, our method is easy to implement and deploy; over 20$\times$ speedup and 90% energy savings have been observed compared to highly optimized baselines in MKLDNN and CUDNN on NVIDIA Jetson Nano. Consequently, our approach opens up a new direction of research with a huge potential for on-device training.

对象检测是计算机视觉中的重要下游任务。对于车载边缘计算平台，很难实现实时检测要求。而且，由大量可分开的卷积层建立的轻巧模型无法达到足够的精度。我们引入了一种新的轻质卷积技术GSCONV，以减轻模型，但保持准确性。 GSCONV在模型的准确性和速度之间取得了极好的权衡。而且，我们提供了一个设计范式，即纤细的颈部，以实现探测器的更高计算成本效益。在二十多组比较实验中，我们的方法的有效性得到了强有力的证明。特别是，通过我们的方法改善的检测器获得了最先进的结果（例如，与原件相比，在Tesla T4 GPU上以〜100fps的速度为70.9％MAP0.5。代码可从https://github.com/alanli1997/slim-neck-by-gsconv获得。

Deploying convolutional neural networks (CNNs) on embedded devices is difficult due to the limited memory and computation resources. The redundancy in feature maps is an important characteristic of those successful CNNs, but has rarely been investigated in neural architecture design. This paper proposes a novel Ghost module to generate more feature maps from cheap operations. Based on a set of intrinsic feature maps, we apply a series of linear transformations with cheap cost to generate many ghost feature maps that could fully reveal information underlying intrinsic features. The proposed Ghost module can be taken as a plug-and-play component to upgrade existing convolutional neural networks. Ghost bottlenecks are designed to stack Ghost modules, and then the lightweight Ghost-Net can be easily established. Experiments conducted on benchmarks demonstrate that the proposed Ghost module is an impressive alternative of convolution layers in baseline models, and our GhostNet can achieve higher recognition performance (e.g. 75.7% top-1 accuracy) than MobileNetV3 with similar computational cost on the ImageNet ILSVRC-2012 classification dataset. Code is available at https: //github.com/huawei-noah/ghostnet.

这项工作提出了一种在像素处理器阵列（PPA）传感器上实施完全卷积神经网络（FCN）的方法，并演示了粗分割和对象定位任务。我们使用batchnorm，小组卷积和可学习的二进制阈值设计和训练二进制的FCN，用于二进制重量和激活，生产足够小的网络，以嵌入PPA的焦平面上，并具有有限的本地记忆资源，并使用平行基本添加添加添加的网络/仅减去，转移和位操作。我们演示了PPA设备上FCN的首次实现，并在像素级处理器中完全执行三个卷积层。我们使用此体系结构来证明使用SCAMP-5 PPA视觉芯片在280 fps处的对象分割和定位的推理生成热图。

由于存储器和计算资源有限，部署在移动设备上的卷积神经网络（CNNS）是困难的。我们的目标是通过利用特征图中的冗余来设计包括CPU和GPU的异构设备的高效神经网络，这很少在神经结构设计中进行了研究。对于类似CPU的设备，我们提出了一种新颖的CPU高效的Ghost（C-Ghost）模块，以生成从廉价操作的更多特征映射。基于一组内在的特征映射，我们使用廉价的成本应用一系列线性变换，以生成许多幽灵特征图，可以完全揭示内在特征的信息。所提出的C-Ghost模块可以作为即插即用组件，以升级现有的卷积神经网络。 C-Ghost瓶颈旨在堆叠C-Ghost模块，然后可以轻松建立轻量级的C-Ghostnet。我们进一步考虑GPU设备的有效网络。在建筑阶段的情况下，不涉及太多的GPU效率（例如，深度明智的卷积），我们建议利用阶段明智的特征冗余来制定GPU高效的幽灵（G-GHOST）阶段结构。舞台中的特征被分成两个部分，其中使用具有较少输出通道的原始块处理第一部分，用于生成内在特征，另一个通过利用阶段明智的冗余来生成廉价的操作。在基准测试上进行的实验证明了所提出的C-Ghost模块和G-Ghost阶段的有效性。 C-Ghostnet和G-Ghostnet分别可以分别实现CPU和GPU的准确性和延迟的最佳权衡。代码可在https://github.com/huawei-noah/cv-backbones获得。

我们日常生活中的深度学习是普遍存在的，包括自驾车，虚拟助理，社交网络服务，医疗服务，面部识别等，但是深度神经网络在训练和推理期间需要大量计算资源。该机器学习界主要集中在模型级优化（如深度学习模型的架构压缩），而系统社区则专注于实施级别优化。在其间，在算术界中提出了各种算术级优化技术。本文在模型，算术和实施级技术方面提供了关于资源有效的深度学习技术的调查，并确定了三种不同级别技术的资源有效的深度学习技术的研究差距。我们的调查基于我们的资源效率度量定义，阐明了较低级别技术的影响，并探讨了资源有效的深度学习研究的未来趋势。

In this paper we describe a new mobile architecture, MobileNetV2, that improves the state of the art performance of mobile models on multiple tasks and benchmarks as well as across a spectrum of different model sizes. We also describe efficient ways of applying these mobile models to object detection in a novel framework we call SSDLite. Additionally, we demonstrate how to build mobile semantic segmentation models through a reduced form of DeepLabv3 which we call Mobile DeepLabv3.is based on an inverted residual structure where the shortcut connections are between the thin bottleneck layers. The intermediate expansion layer uses lightweight depthwise convolutions to filter features as a source of non-linearity. Additionally, we find that it is important to remove non-linearities in the narrow layers in order to maintain representational power. We demonstrate that this improves performance and provide an intuition that led to this design.Finally, our approach allows decoupling of the input/output domains from the expressiveness of the transformation, which provides a convenient framework for further analysis. We measure our performance on ImageNet [1] classification, COCO object detection [2], VOC image segmentation [3]. We evaluate the trade-offs between accuracy, and number of operations measured by multiply-adds (MAdd), as well as actual latency, and the number of parameters.

深神网络的对象探测器正在不断发展，并用于多种应用程序，每个应用程序都有自己的要求集。尽管关键安全应用需要高准确性和可靠性，但低延迟任务需要资源和节能网络。不断提出了实时探测器，在高影响现实世界中是必需的，但是它们过分强调了准确性和速度的提高，而其他功能（例如多功能性，鲁棒性，资源和能源效率）则被省略。现有网络的参考基准不存在，设计新网络的标准评估指南也不存在，从而导致比较模棱两可和不一致的比较。因此，我们对广泛的数据集进行了多个实时探测器（基于锚点，关键器和变压器）的全面研究，并报告了一系列广泛指标的结果。我们还研究了变量，例如图像大小，锚固尺寸，置信阈值和架构层对整体性能的影响。我们分析了检测网络的鲁棒性，以防止分配变化，自然腐败和对抗性攻击。此外，我们提供了校准分析来评估预测的可靠性。最后，为了强调现实世界的影响，我们对自动驾驶和医疗保健应用进行了两个独特的案例研究。为了进一步衡量关键实时应用程序中网络的能力，我们报告了在Edge设备上部署检测网络后的性能。我们广泛的实证研究可以作为工业界对现有网络做出明智选择的指南。我们还希望激发研究社区的设计和评估网络的新方向，该网络着重于更大而整体的概述，以实现深远的影响。

已经提出了高效和自适应计算机视觉系统以使计算机视觉任务，例如图像分类和对象检测，针对嵌入或移动设备进行了优化。这些解决方案最近的起源，专注于通过设计具有近似旋钮的自适应系统来优化模型（深神经网络，DNN）或系统。尽管最近的几项努力，但我们表明现有解决方案遭受了两个主要缺点。首先，系统不考虑模型的能量消耗，同时在制定要运行的模型的决定时。其次，由于其他共同居民工作负载，评估不考虑设备上的争用的实际情况。在这项工作中，我们提出了一种高效和自适应的视频对象检测系统，这是联合优化的精度，能量效率和延迟。底层Virtuoso是一个多分支执行内核，它能够在精度 - 能量 - 延迟轴上的不同运行点处运行，以及轻量级运行时调度程序，以选择最佳的执行分支以满足用户要求。要与Virtuoso相当比较，我们基准于15件最先进的或广泛使用的协议，包括更快的R-CNN（FRCNN），YOLO V3，SSD，培训台，SELSA，MEGA，REPP，FastAdapt和我们的内部FRCNN +，YOLO +，SSD +和高效+（我们的变体具有增强的手机效率）的自适应变体。通过这种全面的基准，Virtuoso对所有上述协议显示出优势，在NVIDIA Jetson Mobile GPU上的每一项效率水平上引领精度边界。具体而言，Virtuoso的准确性为63.9％，比一些流行的物体检测模型高于10％，51.1％，yolo为49.5％。

用于移动设备的有效神经网络骨干通常针对诸如FLOPS或参数计数之类的指标进行优化。但是，这些指标在移动设备上部署时可能与网络的延迟不太相关。因此，我们通过在移动设备上部署多个移动友好网络来对不同指标进行广泛的分析。我们在最近有效的神经网络中识别和分析建筑和优化瓶颈，并提供减轻这些瓶颈的方法。为此，我们设计了一个高效的骨干莫比尼蛋白，在iPhone12上的推理时间低于1毫秒，ImageNet上的Top-1精度为75.9％。我们表明，Mobileone在高效体系结构中实现了最先进的性能，同时在移动设备上的速度更快。我们的最佳模型在38倍的速度中，在Imagenet上的性能与移动形式相似。与在类似延迟时，我们的模型在ImageNet上获得了2.3％的TOP-1精度。此外，我们表明我们的模型概括为多个任务 - 图像分类，对象检测和语义分割，与在移动设备上部署时现有的有效体系结构相比，延迟和准确性的显着提高。

Although RGB-D sensors have enabled major breakthroughs for several vision tasks, such as 3D reconstruction, we have not attained the same level of success in highlevel scene understanding. Perhaps one of the main reasons is the lack of a large-scale benchmark with 3D annotations and 3D evaluation metrics. In this paper, we introduce an RGB-D benchmark suite for the goal of advancing the state-of-the-arts in all major scene understanding tasks. Our dataset is captured by four different sensors and contains 10,335 RGB-D images, at a similar scale as PASCAL VOC. The whole dataset is densely annotated and includes 146,617 2D polygons and 64,595 3D bounding boxes with accurate object orientations, as well as a 3D room layout and scene category for each image. This dataset enables us to train data-hungry algorithms for scene-understanding tasks, evaluate them using meaningful 3D metrics, avoid overfitting to a small testing set, and study cross-sensor bias. Testing Set Training Set NYU (795 images) SUN RGB-D (5,285 images) NYU 32.50 34.33 SUN RGB-D 15.78 33.20

Many state-of-the-art deep learning models for computer vision tasks are based on the transformer architecture. Such models can be computationally expensive and are typically statically set to meet the deployment scenario. However, in real-time applications, the resources available for every inference can vary considerably and be smaller than what state-of-the-art models use. We can use dynamic models to adapt the model execution to meet real-time application resource constraints. While prior dynamic work has primarily minimized resource utilization for less complex input images while maintaining accuracy and focused on CNNs and early transformer models such as BERT, we adapt vision transformers to meet system dynamic resource constraints, independent of the input image. We find that unlike early transformer models, recent state-of-the-art vision transformers heavily rely on convolution layers. We show that pretrained models are fairly resilient to skipping computation in the convolution and self-attention layers, enabling us to create a low-overhead system for dynamic real-time inference without additional training. Finally, we create a optimized accelerator for these dynamic vision transformers in a 5nm technology. The PE array occupies 2.26mm$^2$ and is 17 times faster than a NVIDIA TITAN V GPU for state-of-the-art transformer-based models for semantic segmentation.

Convolutional neural networks (CNN) have recently shown outstanding image classification performance in the largescale visual recognition challenge (ILSVRC2012). The success of CNNs is attributed to their ability to learn rich midlevel image representations as opposed to hand-designed low-level features used in other image classification methods. Learning CNNs, however, amounts to estimating millions of parameters and requires a very large number of annotated image samples. This property currently prevents application of CNNs to problems with limited training data.In this work we show how image representations learned with CNNs on large-scale annotated datasets can be efficiently transferred to other visual recognition tasks with limited amount of training data. We design a method to reuse layers trained on the ImageNet dataset to compute mid-level image representation for images in the PASCAL VOC dataset. We show that despite differences in image statistics and tasks in the two datasets, the transferred representation leads to significantly improved results for object and action classification, outperforming the current state of the art on Pascal VOC 2007 and 2012 datasets. We also show promising results for object and action localization.

随着深度学习的出现，估计来自单个RGB图像的深度最近受到了很多关注，能够赋予许多不同的应用，从用于计算电影的机器人的路径规划范围。尽管如此，虽然深度地图完全可靠，但对象不连续的估计仍然远非令人满意。这可以有助于卷积运营商自然地聚集在对象不连续性的特征的事实中，导致平滑的过渡而不是明确的边界。因此，为了规避这个问题，我们提出了一种新颖的卷积运营商，明确地定制，以避免不同对象部件的特征聚合。特别地，我们的方法基于借助于超像素估计每个部分深度值。所提出的卷积运算符，我们将“实例卷积”，然后仅在估计的超像素的基础上单独考虑每个对象部分。我们对NYUV2以及IBIMS数据集的评估清楚地展示了在估计遮挡边界周围估算深度的经典卷积上的实例卷积的优越性，同时在其他地方产生了可比结果。代码将在接受时公开提供。

Image segmentation is a key topic in image processing and computer vision with applications such as scene understanding, medical image analysis, robotic perception, video surveillance, augmented reality, and image compression, among many others. Various algorithms for image segmentation have been developed in the literature. Recently, due to the success of deep learning models in a wide range of vision applications, there has been a substantial amount of works aimed at developing image segmentation approaches using deep learning models. In this survey, we provide a comprehensive review of the literature at the time of this writing, covering a broad spectrum of pioneering works for semantic and instance-level segmentation, including fully convolutional pixel-labeling networks, encoder-decoder architectures, multi-scale and pyramid based approaches, recurrent networks, visual attention models, and generative models in adversarial settings. We investigate the similarity, strengths and challenges of these deep learning models, examine the most widely used datasets, report performances, and discuss promising future research directions in this area.

This paper addresses the problem of estimating the depth map of a scene given a single RGB image. We propose a fully convolutional architecture, encompassing residual learning, to model the ambiguous mapping between monocular images and depth maps. In order to improve the output resolution, we present a novel way to efficiently learn feature map up-sampling within the network. For optimization, we introduce the reverse Huber loss that is particularly suited for the task at hand and driven by the value distributions commonly present in depth maps. Our model is composed of a single architecture that is trained end-to-end and does not rely on post-processing techniques, such as CRFs or other additional refinement steps. As a result, it runs in real-time on images or videos. In the evaluation, we show that the proposed model contains fewer parameters and requires fewer training data than the current state of the art, while outperforming all approaches on depth estimation. Code and models are publicly available 5 .