Phoenixtree_DongZhao

MyDLNote - Attention: [NLA系列] Asymmetric Non-local Neural Networks for Semantic Segmentation

Asymmetric Non-local Neural Networks for Semantic Segmentation

Zhen Zhu , Mengde Xu , Song Bai , Tengteng Huang , Xiang Bai

Huazhong University of Science and Technology, University of Oxford

[GitHub]： https://github.com/MendelXu/ANN

[paper]： https://arxiv.org/pdf/1908.07678.pdf

这篇文章的写作很棒。

[Non-Local Attention 系列]

Non-local neural networks

GCNet: Non-local Networks Meet Squeeze-Excitation Networks and Beyond [my CSDN]

Asymmetric Non-local Neural Networks for Semantic Segmentation [my CSDN]

Efficient Attention: Attention with Linear Complexities [my CSDN]

CCNet: Criss-Cross Attention for Semantic Segmentation [my CSDN]

Non-locally Enhanced Encoder-Decoder Network for Single Image De-raining [my CSDN]

Image Restoration via Residual Non-local Attention Networks [my CSDN]

Table of Contents

Asymmetric Non-local Neural Networks for Semantic Segmentation

Abstract

Introduction

Asymmetric Non-local Neural Network

Revisiting Non-local Block

Asymmetric Pyramid Non-local Block

Asymmetric Fusion Non-local Block

Network Architecture

Experiments

Datasets and Evaluation Metrics

Implementation Details

Comparisons with Other Methods

Conclusion

Abstract

The non-local module works as a particularly useful technique for semantic segmentation while criticized for its prohibitive computation and GPU memory occupation.

[文章的motivation，即文章要解决的问题]

In this paper, we present Asymmetric Non-local Neural Network to semantic segmentation, which has two prominent components: Asymmetric Pyramid Non-local Block (APNB) and Asymmetric Fusion Non-local Block (AFNB). APNB leverages a pyramid sampling module into the non-local block to largely reduce the computation and memory consumption without sacrificing the performance. AFNB is adapted from APNB to fuse the features of different levels under a sufficient consideration of long range dependencies and thus considerably improves the performance.

[网络结构介绍，简洁而清晰]

Extensive experiments on semantic segmentation benchmarks demonstrate the effectiveness and efficiency of our work. In particular, we report the state-of-the-art performance of 81.3 mIoU on the Cityscapes test set. For a 256 × 128 input, APNB is around 6 times faster than a non-local block on GPU while 28 times smaller in GPU running memory occupation.

[最后，实验结果要和motivation相呼应，即说明文章提出的方法确实有效解决了motivation提出的问题]

Introduction

Some recent studies [20, 33, 47] indicate that the performance could be improved if making sufficient use of long range dependencies. However, models that solely rely on convolutions exhibit limited ability in capturing these long range dependencies. A possible reason is the receptive field of a single convolutional layer is inadequate to cover correlated areas. Choosing a big kernel or composing a very deep network is able to enlarge the receptive field. However, such strategies require extensive computation and parameters, thus being very inefficient [44]. Consequently, several works [33, 47] resort to use global operations like non-local means [2] and spatial pyramid pooling [12, 16].

传统CNN不能高效的捕捉long range dependency. Non-local 和 spatial pyramid pooling 可以有效解决这个问题。

In [33], Wang et al. combined CNNs and traditional non-local means [2] to compose a network module named nonlocal block in order to leverage features from all locations in an image. This module improves the performance of existing methods [33]. However, the prohibitive computational cost and vast GPU memory occupation hinder its usage in many real applications. The architecture of a common non-local block [33] is depicted in Fig. 1(a). The block first calculates the similarities of all locations between each other, requiring a matrix multiplication of computational complexity , given an input feature map with size $C\times H\times W$ . Then it requires another matrix multiplication of computational complexity to gather the influence of all locations to themselves. Concerning the high complexity brought by the matrix multiplications, we are interested in this work if there are efficient ways to solve this without sacrificing the performance.

描述了Non-local的结构和复杂度。

这两段是问题的提出。

Figure 1: Architecture of a standard non-local block (a) and the asymmetric non-local block (b). $N=H\cdot W$ while $S\ll N$ .

We notice that as long as the outputs of the key branch and value branch hold the same size, the output size of the non-local block remains unchanged. Considering this, if we could sample only a few representative points from key branch and value branch, it is possible that the time complexity is significantly decreased without sacrificing the performance. This motivation is demonstrated in Fig. 1 when changing a large value N in the key branch and value branch to a much smaller value S (From (a) to (b)).

本文的观察和动机。Non-local是计算每个点和所有点的相关性（即每个像素点都对应一张attention map）；而本文认为不需要计算每个点的attention map，而是计算每个像素点与某几个局部区域的 attention map。

In this paper, we propose a simple yet effective nonlocal module called Asymmetric Pyramid Non-local Block (APNB) to decrease the computation and GPU memory consumption of the standard non-local module [33] with applications to semantic segmentation. Motivated by the spatial pyramid pooling [12, 16, 47] strategy, we propose to embed a pyramid sampling module into non-local blocks, which could largely reduce the computation overhead of matrix multiplications yet provide substantial semantic feature statistics. This spirit is also related to the sub-sampling tricks [33] (e.g., max pooling). Our experiments suggest that APNB yields much better performance than those sub-sampling tricks with a decent decrease of computations. To better illustrate the boosted efficiency, we compare the GPU times of APNB and a standard non-local block in Fig. 2, averaging the running time of 10 different runs with the same configuration. Our APNB largely reduces the time cost on matrix multiplications, thus being nearly 6 times faster than a non-local block.

描述APNB的形成机制（Non-Local + Spatial Pyramid Pooling），具备低复杂度、高 performance。

Besides, we also adapt APNB to fuse the features of different stages of a deep network, which brings a considerable improvement over the baseline model. We call the adapted block as Asymmetric Fusion Non-local Block (AFNB). AFNB calculates the correlations between every pixel of the low-level and high-level feature maps, yielding a fused feature with long range interactions. Our network is built based on a standard ResNet-FCN model by integrating APNB and AFNB together.

描述AFNB：融合了low-level和high-level的特征，生成了具有long range interactions的、融合的特征。

Recent advances focus on exploring the context information and can be roughly categorized into five directions:

为了挖掘 context information，现有的方法大致包括5个研究方向：

Encoder-Decoder：U-Net

Conditional Random Field: Deeplab

Different Convolutions: Dilated Conv.

Spatial Pyramid Pooling: PSPNet, Atrous Spatial Pyramid Pooling layer (ASPP)

Non-local Network: GCNet(19CVPR), NLNet

Different from these works, our network uniquely incorporates pyramid sampling strategies with non-local blocks to capture the semantic statistics of different scales with only a minor budget of computation, while maintaining the excellent performance as the original non-local modules.

Asymmetric Non-local Neural Network

While APNB aims to decrease the computational overhead of non-local blocks, AFNB improves the learning capacity of non-local blocks thereby improving the segmentation performance.

APNB的目标是降低 non-local blocks 的计算开销，而AFNB提高了 non-local blocks 的学习能力，从而提高了分割性能。

Revisiting Non-local Block

Non-local 原理就不多写了，这里列出后面要用到的公式。函数can take the form from softmax, rescaling, and none. 选择 softmax，就是Embedded Gaussian.

Asymmetric Pyramid Non-local Block

Motivation and Analysis

By inspecting the general computing flow of a non-local block, one could clearly find that Eq. (2) and Eq. (4) dominate the computation. The time complexities of the two matrix multiplications are both $\small O(\hat{C}N^2)=O(\hat{C}H^2W^2)$ . In semantic segmentation, the output of the network usually has a large resolution to retain detailed semantic features [6, 47]. That means N is large (for example in our training phase, N = 96 × 96 = 9216). Hence, the large matrix multiplication is the main cause of the inefficiency of a non-local block (see our statistic in Fig. 2). A more straightforward pipeline is given as

这段很好理解，就是详细介绍 Non-local 的复杂度为啥那么高。

We hold a key yet intuitive observation that by changing N to another number $\small S(S\ll N)$ , the output size will remain the same, as

Returning to the design of the non-local block, changing N to a small number S is equivalent to sampling several representative points from θ and γ instead of feeding all the spatial points, as illustrated in Fig. 1. Consequently, the computational complexity could be considerably decreased.

这段也很好理解，就是把N缩小到S，S远远小于N。那这个操作是怎么实现的呢？

Solution

Based on the above observation, we propose to add sampling modules $\small \mathcal{P}_\theta$ and $\small \mathcal{P}_\gamma$ after θ and γ to sample several sparse anchor points denoted as $\small \theta _P\in\mathcal{R}^{\hat{C}\times S}$ and $\small \gamma_P\in\mathcal{R}^{\hat{C}\times S}$ , where S is the number of sampled anchor points. Mathematically, this is computed by

$\large \theta _P=\mathcal{P}_\theta (\theta ),~\gamma _P=\mathcal{P}_\gamma(\gamma )$ (8)

The similarity matrix $\small {V}_P$ between $\small \theta$ and the anchor points $\small {\theta}_P$ is thus calculated by

Note that $\small {V}_P$ is an asymmetric matrix of size N × S. $\small {V}_P$ then goes through the same normalizing function as a standard non-local block, giving the unified similarity matrix $\small \vec{V}_P$ . And the attention output is acquired by

where the output is in the same size as that of Eq. (4). Following non-local blocks, the final output $\small Y_P$ is given as

The time complexity of such an asymmetric matrix multiplication is only O(CNS ˆ ), significantly lower than O(CNˆ 2 ) in a standard non-local block. It is ideal that S should be much smaller than N. However, it is hard to ensure that when S is small, the performance would not drop too much in the meantime.

这部分详细介绍采用pyramid pooling降低复杂度。Non-local 中，输入的 query 和 key 是经过 1x1 卷积得到的 CxN 维；而这里，query 和 key 是经过 1x1 卷积，还要用 pyramid pooling 进行采样，得到 CxS 维。即，non-local是把每个像素点作为 query，而APNB是对图像做采样，用几个 sparse anchor points 作为query。如此，计算复杂度从 $\small O(\hat{C}N^2)$ 降低到 $O(\hat{C}NS)$ 。

作者同时认为，S 不能太小。

下面部分则是更进一步介绍具体的 pyramid pooling 是如何操作的。

Spatial Pyramid Pooling

As discovered by previous works [16, 47], global and multi-scale representations are useful for categorizing scene semantics. Such representations can be comprehensively carved by Spatial Pyramid Pooling [16], which contains several pooling layers with different output sizes in parallel. In addition to this virtue, spatial pyramid pooling is also parameter-free and very efficient. Therefore, we embed pyramid pooling in the non-local block to enhance the global representations while reducing the computational overhead.

By doing so, we now arrive at the final formulation of Asymmetric Pyramid Non-local Block (APNB), as given in Fig. 3. As can be seen, our APNB derives from the design of a standard non-local block [33]. A vital change is to add a spatial pyramid pooling module after θ and γ respectively to sample representative anchors. This sampling process is clearly depicted in Fig. 4, where several pooling layers are applied after θ or γ and then the four pooling results are flattened and concatenated to serve as the input to the next layer. We denote the spatial pyramid pooling modules as $\small \mathcal{P}^n_\theta$ and $\small \mathcal{P}^n_\gamma$ , where the superscript $\small n$ means the width (or height) of the output size of the pooling layer (empirically, the width is equal to the height). In our model, we set $\small n\subseteq \{1,3,6,8\}$ . Then the total number of the anchor points is

Figure 3: Overview of the proposed Asymmetric Non-local Neural Network. In our implementation, the key branch and the value branch in APNB share the same 1×1 convolution and sampling module, which decreases the number of parameters and computation without sacrificing the performance.

Figure 4: Demonstration of the pyramid max or average sampling process.

SPP具体的操作细节。以H = 128 and W = 256为例，本文的方法降低复杂度为non-local的 (256x128)/112=298倍。

注意：在APNB中，作者把key和value的conv1x1及pyramid pooling共享了，即用的是同一组矩阵。

Asymmetric Fusion Non-local Block

Fusing features of different levels are helpful to semantic segmentation and object tracking as hinted in [16, 18, 26, 41, 46, 51]. Common fusing operations such as addition/concatenation, are conducted in a pixel-wise and local manner. We provide an alternative that leverages long range dependencies through a non-local block to fuse multi-level features, called Fusion Non-local Block. A standard non-local block only has one input source while FNB has two: a high-level feature map $\small X_h\in \mathcal{R}^{C_h\times N_h}$ and a low-level feature map $\small X_l\in \mathcal{R}^{C_l\times N_l}$ . Nh and Nl are the numbers of spatial locations of $\small X_h$ and $\small X_l$ , respectively. $\small C_h$ and $\small C_l$ are the channel numbers of $\small X_h$ and $\small X_l$ , respectively. Likewise, 1 × 1 convolutions $\small W_h^\varphi$ , $\small W_l^\theta$ and $\small W_l^\gamma$ are used to transform $\small X_h$ and $\small X_l$ to embeddings $\small \varphi_h\in \mathcal{R}^{\hat{C}\times N_h}$ , $\small \theta _l\in \mathcal{R}^{\hat{C}\times N_l}$ and $\small \gamma _l\in \mathcal{R}^{\hat{C}\times N_l}$ as

the similarity matrix $\small V_F$

The output $\small O_F\in \mathcal{R}^{N_l\times \hat{C}}$ reflects the bonus of $\small X_l$ to $\small X_h$ , which are carefully selected from all locations in $\small X_l$ .

AFNB的细节，其实图3已经解释的很清楚了。

Network Architecture

The overall architecture of our network is depicted in Fig. 3. We choose ResNet-101 [13] as our backbone network following the choice of most previous works [38, 47, 48]. We remove the last two down-sampling operations and use the dilation convolutions instead to hold the feature maps from the last two stages 1/8 of the input image. Concretely, all the feature maps in the last three stages have the same spatial size. According to our experimental trials, we fuse the features of Stage4 and Stage5 using AFNB. The fused features are thereupon concatenated with the feature maps after Stage5, avoiding situations that AFNB could not produce accurate strengthened features particularly when the training just begins and degrades the overall performance. Such features, full of rich long range cues from different feature levels, serve as the input to APNB, which then help to discover the correlations among pixels. As done for AFNB, the output of APNB is also concatenated with its input source. Note that in our implementation for APNB, $\small W_\theta$ and $\small W_\gamma$ share parameters in order to save parameters and computation, following the design of [42]. This design doesn’t decrease the performance of APNB. Finally, a classifier is followed up to produce channel-wise semantic maps that later receive their supervisions from the ground truth maps. Note we also add another supervision to Stage4 following the settings of [47], as it is beneficial to improve the performance.

网络构成细节：

1. ResNet 101

2. stage 4 和 5 分辨率不变，采用的正是 dilated ResNet [参考我的博客 Dilated Residual Networks]

3. AFNB 融合的是 stage 4 和 5

4. AFNB 之后跟 APNB，残差连接方式

5. $\small W_\theta$ and $\small W_\gamma$ share parameters [42: OCnet: Object context network for scene parsing]

6. add another supervision to Stage4 [47: Pyramid scene parsing network]

Experiments

Datasets and Evaluation Metrics

Datasets：Cityscapes [9], ADE20K [50] and PASCAL Context [21].

Evaluation Metric：Mean IoU (mean of classwise intersection over union)

Implementation Details

Training Objectives

Following [47 Pyramid scene parsing network], our model has two supervisions: one after the final output of our model while another at the output layer of Stage4. For Lfinal, we perform online hard pixel mining, which excels at coping with difficult cases.

Comparisons with Other Methods

1. Efficiency Comparison with Non-local Block

2. Performance Comparisons

3. Ablation Study

Efficacy of the APNB and AFNB

Selection of Sampling Methods：

Influence of the Anchor Points Numbers

Conclusion

个人总结：

这篇文章两个亮点：

1. 与Non-local 不同的是：

Non-local：每个点（query）与每个点之间的相关；pixel-wise

APNB：每个点与局部内容之间的关系；patch-wise

2. 融合不同level的的特征：AFNB

你可能感兴趣的:(deep,learning,semantic,segmentation,non-local)

机器学习与深度学习间关系与区别 ℒℴѵℯ心·动ꦿ໊ོ꫞ 人工智能学习深度学习 python
一、机器学习概述定义机器学习（MachineLearning,ML）是一种通过数据驱动的方法，利用统计学和计算算法来训练模型，使计算机能够从数据中学习并自动进行预测或决策。机器学习通过分析大量数据样本，识别其中的模式和规律，从而对新的数据进行判断。其核心在于通过训练过程，让模型不断优化和提升其预测准确性。主要类型1.监督学习（SupervisedLearning）监督学习是指在训练数据集中包含输入
JavaScript 中，深拷贝（Deep Copy）和浅拷贝（Shallow Copy）跳房子的前端前端面试 javascript 开发语言 ecmascript
在JavaScript中，深拷贝（DeepCopy）和浅拷贝（ShallowCopy）是用于复制对象或数组的两种不同方法。了解它们的区别和应用场景对于避免潜在的bugs和高效地处理数据非常重要。以下是对深拷贝和浅拷贝的详细解释，包括它们的概念、用途、优缺点以及实现方式。1.浅拷贝（ShallowCopy）概念定义：浅拷贝是指创建一个新的对象或数组，其中包含了原对象或数组的基本数据类型的值和对引用数
深度 Qlearning：在直播推荐系统中的应用 AGI通用人工智能之禅程序员提升自我硅基计算碳基计算认知计算生物计算深度学习神经网络大数据 AIGC AGI LLM Java Python 架构设计 Agent 程序员实现财富自由
深度Q-learning：在直播推荐系统中的应用关键词：深度Q-learning,强化学习,直播推荐系统,个性化推荐1.背景介绍1.1问题的由来随着互联网技术的飞速发展,直播平台如雨后春笋般涌现。面对海量的直播内容,用户很难快速找到自己感兴趣的内容。因此,个性化推荐系统在直播平台中扮演着越来越重要的角色。1.2研究现状目前,主流的个性化推荐算法包括协同过滤、基于内容的推荐等。这些方法在一定程度上缓
深度学习-点击率预估-研究论文2024-09-14速读 sp_fyf_2024 深度学习人工智能
深度学习-点击率预估-研究论文2024-09-14速读1.DeepTargetSessionInterestNetworkforClick-ThroughRatePredictionHZhong,JMa,XDuan,SGu,JYao-2024InternationalJointConferenceonNeuralNetworks,2024深度目标会话兴趣网络用于点击率预测摘要：这篇文章提出了一种新
探索未来，大规模分布式深度强化学习——深入解析IMPALA架构汤萌妮Margaret
探索未来，大规模分布式深度强化学习——深入解析IMPALA架构scalable_agent项目地址:https://gitcode.com/gh_mirrors/sc/scalable_agent在当今的人工智能研究前沿，深度强化学习（DRL）因其在复杂任务中的卓越表现而备受瞩目。本文要介绍的是一个开源于GitHub的重量级项目：“ScalableDistributedDeep-RLwithImp
云服务业界动态简报-20180128 Captain7
一、青云青云QingCloud推出深度学习平台DeepLearningonQingCloud，包含了主流的深度学习框架及数据科学工具包，通过QingCloudAppCenter一键部署交付，可以让算法工程师和数据科学家快速构建深度学习开发环境，将更多的精力放在模型和算法调优。二、腾讯云1.腾讯云正式发布腾讯专有云TCE(TencentCloudEnterprise)矩阵，涵盖企业版、大数据版、AI
机器学习VS深度学习 nfgo 机器学习
机器学习（MachineLearning,ML）和深度学习（DeepLearning,DL）是人工智能（AI）的两个子领域，它们有许多相似之处，但在技术实现和应用范围上也有显著区别。下面从几个方面对两者进行区分：1.概念层面机器学习：是让计算机通过算法从数据中自动学习和改进的技术。它依赖于手动设计的特征和数学模型来进行学习，常用的模型有决策树、支持向量机、线性回归等。深度学习：是机器学习的一个子领
ResNet的半监督和半弱监督模型 Valar_Morghulis
Billion-scalesemi-supervisedlearningforimageclassificationhttps://arxiv.org/pdf/1905.00546.pdfhttps://github.com/facebookresearch/semi-supervised-ImageNet1K-models/权重在timm中也有：https://hub.fastgit.org/r
联邦学习 Federated learning Google I/O‘19 笔记努力搬砖的星期五笔记联邦学习机器学习机器学习 tensorflow
FederatedLearning:MachineLearningonDecentralizeddatahttps://www.youtube.com/watch?v=89BGjQYA0uE文章目录FederatedLearning:MachineLearningonDecentralizeddata1.DecentralizeddataEdgedevicesGboard:mobilekeyboa
PCL 怎样可视化深度图像 LeonDL168 PCL 计算机视觉人工智能视觉检测图像处理算法
本小节讲解如何可视化深度图像的两种方法，在3D视窗中以点云形式进行可视化（深度图像来源于点云），另一种是，将深度值映射为颜色，从而以彩色图像方式可视化深度图像。代码首先，在PCL（PointCloudLearning）中国协助发行的书提供光盘的第7章例2文件夹中，打开名为range_image_visualization.cpp的代码文件，同文件夹下可以找到相关的测试点云文件room_scan1.
el-dialog高度设置夏之小星星前端 vue.js elementui css
el-dialog高度设置::v-deep.el-dialog{height:78vh;overflow:auto;}
elementuiPlus取消el-input的边框 qq_39016177 elementui
elementuiPlus取消el-input的边框1.通常取消边框的方法设置border为none2.还有其他类似边框的例如outlinebox-shadow这两个属性都是会产生边框效果3.el-input需要更改的话–如下需要修改box-shadow为空即可上代码:deep(.el-input__wrapper){align-items:center;background-color:#F7F
[C++11] 移动语意和移动构造函数图王大胜计算机学科基础开发语言 c++移动语意移动构造函数
说明：移动语义（MoveSemantics）是C++11引入的一个重要概念，旨在提高大型对象（特别是那些涉及资源管理的对象）的复制效率。移动语义允许资源从一个对象“移动”到另一个对象，而不是进行昂贵的复制操作。这种机制通过右值引用（right-valuereference）和移动构造函数（moveconstructor）以及移动赋值操作符（moveassignmentoperator）来实现。而移
论文-A Stack-Propagation Framework with Token-Level Intent Detection for Spoken Language Understanding 魏鹏飞
1.简称论文《AStack-PropagationFrameworkwithToken-LevelIntentDetectionforSpokenLanguageUnderstanding》，作者LiboQin(HarbinInstituteofTechnology,China)，经典的NLU论文（SemanticFrame）。2.摘要意图检测和槽位填充是构建口语理解（SLU）系统的两个主要任务。
概率潜在语义分析（Probabilistic Latent Semantic Analysis,PLSA）—无监督学习方法、概率模型、生成模型、共现模型、非线性模型、参数化模型、批量学习剑海风云 Artificial Intelligence 人工智能机器学习概率潜在语义分析 PLSA
定义输入:设单词集合为W={ω1,ω2,⋯ ,ωM}W=\{\omega_1,\omega_2,\cdots,\omega_M\}W={ω1,ω2,⋯,ωM},文本集合为D={d1,d2,⋯ ,dN}D=\{d_1,d_2,\cdots,d_N\}D={d1,d2,⋯,dN},话题集合为Z={z1,z2,⋯ ,zN}Z=\{z_1,z_2,\cdots,z_N\}Z={z1,z2,⋯,zN},共现
【双语新闻】AGI安全与对齐，DeepMind近期工作曲奇人工智能安全 agi 安全 llama 人工智能
我们想与AF社区分享我们最近的工作总结。以下是关于我们正在做什么，为什么会这么做以及我们认为它的意义所在的一些详细信息。我们希望这能帮助人们从我们的工作基础上继续发展，并了解他们的工作如何与我们相关联。byRohinShah,SebFarquhar,AncaDragan21stAug2024AIAlignmentForumWewantedtosharearecapofourrecentoutput
Awesome TensorFlow weixin_30594001 人工智能移动开发大数据
AwesomeTensorFlowAcuratedlistofawesomeTensorFlowexperiments,libraries,andprojects.Inspiredbyawesome-machine-learning.WhatisTensorFlow?TensorFlowisanopensourcesoftwarelibraryfornumericalcomputationusin
【ShuQiHere】探索人工智能核心：机器学习的奥秘 ShuQiHere 人工智能机器学习
【ShuQiHere】什么是机器学习？机器学习（MachineLearning,ML）是人工智能（ArtificialIntelligence,AI）中最关键的组成部分之一。它使得计算机不仅能够处理数据，还能从数据中学习，从而做出预测和决策。无论是语音识别、自动驾驶还是推荐系统，背后都依赖于机器学习模型。机器学习与传统的编程不同，它不再依赖于人类编写的固定规则，而是通过数据自我改进模型，从而更灵活
综述论文“A Survey of Zero-Shot Learning: Settings, Methods, and Applications” 硅谷秋水机器学习机器学习神经网络深度学习
该零样本学习综述，发表于ACMTrans.Intell.Syst.Technol.10,2,Article13(January2019)摘要：大多数机器学习方法着重于对已经在训练中看到其类别的实例进行分类。实际上，许多应用程序需要对实例进行分类，而这些实例的类以前没有见过。零样本学习（Zero-ShotLearning）是一种强大而有前途的学习范例，其中训练实例涵盖的类别与想分类的类别是不相交的。
机器学习 VS 表示学习 VS 深度学习 Efred.D 人工智能机器学习深度学习人工智能
文章目录前言一、机器学习是什么?二、表示学习三、深度学习总结前言本文主要阐述机器学习,表示学习和深度学习的原理和区别.一、机器学习是什么?机器学习(machinelearning),是从有限的数据集中学习到一定的规律,再把学到的规律应用到一些相似的样本集中做预测.机器学习的历史可以追溯到20世纪40年代McCulloch提出的人工神经元网络,目前学界大致把机器学习分为传统机器学习和机器学习两个类别
端到端的自动驾驶论文与代码整理大别山伧父自动驾驶
LearningbyCheatinggithubcodearxivpaperconferenceonrobotlearning最新进展(May2021)Checkoutourlatestfollow-upwork:WorldonRails(2020)Checkoutoursubmissiontothe2020CARLAChallenge!pass
Lt-8 Multithreading yanlingyun0210 java
IntendedLearningOutcomesTounderstandtheconceptofconcurrency.Tounderstandthedifferenceofaprocessandathread.TodefineathreadusingtheThreadclassandRunnableinterface.TocontrolthreadswithvariousThreadmethod
如何使用Pytorch-Metric-Learning？鱼儿也有烦恼 PyTorch pytorch
文章目录如何使用Pytorch-Metric-Learning？1.Pytorch-Metric-Learning库9个模块的功能1.1Sampler模块1.2Miner模块1.3Loss模块1.4Reducer模块1.5Distance模块1.6Regularizer模块1.7Trainer模块1.8Tester模块1.9Utils模块2.如何使用PyTorchMetricLearning库中的
[Kaiming]Delving Deep into Rectifiers: Surpassing Human-Level Performance on ImageNet Classification MTandHJ neural networks
文章目录概主要内容PReLUKaiming初始化ForwardcaseBackwardcaseHeK,ZhangX,RenS,etal.DelvingDeepintoRectifiers:SurpassingHuman-LevelPerformanceonImageNetClassification[C].internationalconferenceoncomputervision,2015:1
深度神经网络详解：原理、架构与应用阿达C 活动 dnn 计算机网络人工智能神经网络机器学习深度学习
深度神经网络（DeepNeuralNetwork，DNN）是机器学习领域中最为重要和广泛应用的技术之一。它模仿人脑神经元的结构，通过多层神经元的连接和训练，能够处理复杂的非线性问题。在图像识别、自然语言处理、语音识别等领域，深度神经网络展示了强大的性能。本文将深入解析深度神经网络的基本原理、常见架构及其实际应用。一、深度神经网络的基本原理1.1神经元和感知器神经元是深度神经网络的基本组成单元。一个
前端开发需要了解的算法知识史努比的大头算法前端
手写深拷贝functiondeepClone(obj){//处理基础数据类型和函数if(obj===null||typeofobj!=='object'){returnobj;}//处理数组if(Array.isArray(obj)){returnobj.map(item=>deepClone(item));}//处理对象constclonedObj={};for(constkeyinobj){i
推荐开源项目：PyTorch-Metric-Learning 潘惟妍
推荐开源项目：PyTorch-Metric-Learningpytorch-metric-learningTheeasiestwaytousedeepmetriclearninginyourapplication.Modular,flexible,andextensible.WritteninPyTorch.项目地址:https://gitcode.com/gh_mirrors/py/pytorc
AI推介-多模态视觉语言模型VLMs论文速览（arXiv方向）：2024.07.25-2024.08.01 小小帅AIGC VLM论文时报人工智能语言模型自然语言处理 VLM 大语言模型计算机视觉视觉语言模型
文章目录～1.PayingMoreAttentiontoImage:ATraining-FreeMethodforAlleviatingHallucinationinLVLMs2.MTA-CLIP:Language-GuidedSemanticSegmentationwithMask-TextAlignment3.MarvelOVD:MarryingObjectRecognitionandVisi
推荐：FastAPI驱动的稳定扩散LLMs演示项目褚知茉Jade
推荐：FastAPI驱动的稳定扩散LLMs演示项目FastAPI-for-Machine-Learning-Live-DemoThisrepositorycontainsthefilestobuildyourveryownAIimagegenerationwebapplication!OutlinedarethecorecomponentsoftheFastAPIwebframework,anda
【python】【Ray的概述】资源存储库 python 开发语言
Overview概述Rayisanopen-sourceunifiedframeworkforscalingAIandPythonapplicationslikemachinelearning.Itprovidesthecomputelayerforparallelprocessingsothatyoudon’tneedtobeadistributedsystemsexpert.Rayminimi
redis学习笔记——不仅仅是存取数据 Everyday都不同 returnSource expire/del incr/lpush 数据库分区 redis
最近项目中用到比较多redis，感觉之前对它一直局限于get/set数据的层面。其实作为一个强大的NoSql数据库产品，如果好好利用它，会带来很多意想不到的效果。（因为我搞java，所以就从jedis的角度来补充一点东西吧。PS：不一定全，只是个人理解，不喜勿喷） 1、关于JedisPool.returnSource(Jedis jeids) 这个方法是从red
SQL性能优化-持续更新中。。。。。。 atongyeye oracle sql
1 通过ROWID访问表--索引你可以采用基于ROWID的访问方式情况,提高访问表的效率, , ROWID包含了表中记录的物理位置信息..ORACLE采用索引(INDEX)实现了数据和存放数据的物理位置(ROWID)之间的联系. 通常索引提供了快速访问ROWID的方法,因此那些基于索引列的查询就可以得到性能上的提高. 2 共享SQL语句--相同的sql放入缓存 3 选择最有效率的表
[JAVA语言]JAVA虚拟机对底层硬件的操控还不完善 comsci JAVA虚拟机
如果我们用汇编语言编写一个直接读写CPU寄存器的代码段，然后利用这个代码段去控制被操作系统屏蔽的硬件资源，这对于JVM虚拟机显然是不合法的，对操作系统来讲，这样也是不合法的，但是如果是一个工程项目的确需要这样做，合同已经签了，我们又不能够这样做，怎么办呢？那么一个精通汇编语言的那种X客，是否在这个时候就会发生某种至关重要的作用呢？ &n
lvs- real 男人50 LVS
#!/bin/bash # # Script to start LVS DR real server. # description: LVS DR real server # #. /etc/rc.d/init.d/functions VIP=10.10.6.252 host='/bin/hostname' case "$1" in sta
生成公钥和私钥 oloz DSA 安全加密
package com.msserver.core.util; import java.security.KeyPair; import java.security.PrivateKey; import java.security.PublicKey; import java.security.SecureRandom; public class SecurityUtil {
UIView 中加入的cocos2d，背景透明 374016526 cocos2d glClearColor
要点是首先pixelFormat:kEAGLColorFormatRGBA8，必须有alpha层才能透明。然后view设置为透明glView.opaque = NO;[director setOpenGLView:glView];[self.viewController.view setBackgroundColor:[UIColor clearColor]];[self.viewControll
mysql常用命令香水浓 mysql
连接数据库 mysql -u troy -ptroy 备份表 mysqldump -u troy -ptroy mm_database mm_user_tbl > user.sql 恢复表（与恢复数据库命令相同） mysql -u troy -ptroy mm_database < user.sql 备份数据库 mysqldump -u troy -ptroy
我的架构经验系列文章 - 后端架构 - 系统层面 agevs JavaScript jquery css html5
系统层面：高可用性所谓高可用性也就是通过避免单独故障加上快速故障转移实现一旦某台物理服务器出现故障能实现故障快速恢复。一般来说，可以采用两种方式，如果可以做业务可以做负载均衡则通过负载均衡实现集群，然后针对每一台服务器进行监控，一旦发生故障则从集群中移除；如果业务只能有单点入口那么可以通过实现Standby机加上虚拟IP机制，实现Active机在出现故障之后虚拟IP转移到Standby的快速
利用ant进行远程tomcat部署 aijuans tomcat
在javaEE项目中，需要将工程部署到远程服务器上，如果部署的频率比较高，手动部署的方式就比较麻烦，可以利用Ant工具实现快捷的部署。这篇博文详细介绍了ant配置的步骤（http://www.cnblogs.com/GloriousOnion/archive/2012/12/18/2822817.html），但是在tomcat7以上不适用，需要修改配置，具体如下： 1.配置tomcat的用户角色
获取复利总收入 baalwolf 获取
public static void main(String args[]){ int money=200; int year=1; double rate=0.1; &
eclipse.ini解释 BigBird2012 eclipse
大多数java开发者使用的都是eclipse，今天感兴趣去eclipse官网搜了一下eclipse.ini的配置，供大家参考，我会把关键的部分给大家用中文解释一下。还是推荐有问题不会直接搜谷歌，看官方文档，这样我们会知道问题的真面目是什么，对问题也有一个全面清晰的认识。 Overview 1、Eclipse.ini的作用 Eclipse startup is controlled by th
AngularJS实现分页功能 bijian1013 JavaScript AngularJS 分页
对于大多数web应用来说显示项目列表是一种很常见的任务。通常情况下，我们的数据会比较多，无法很好地显示在单个页面中。在这种情况下，我们需要把数据以页的方式来展示，同时带有转到上一页和下一页的功能。既然在整个应用中这是一种很常见的需求，那么把这一功能抽象成一个通用的、可复用的分页（Paginator）服务是很有意义的。 &nbs
[Maven学习笔记三]Maven archetype bit1129 ArcheType
archetype的英文意思是原型，Maven archetype表示创建Maven模块的模版，比如创建web项目，创建Spring项目等等. mvn archetype提供了一种命令行交互式创建Maven项目或者模块的方式， mvn archetype 1.在LearnMaven-ch03目录下，执行命令mvn archetype:gener
【Java命令三】jps bit1129 Java命令
jps很简单，用于显示当前运行的Java进程，也可以连接到远程服务器去查看 [hadoop@hadoop bin]$ jps -help usage: jps [-help] jps [-q] [-mlvV] [<hostid>] Definitions: <hostid>: <hostname>[:
ZABBIX2.2 2.4 等各版本之间的兼容性 ronin47
zabbix更新很快，从2009年到现在已经更新多个版本，为了使用更多zabbix的新特性，随之而来的便是升级版本，zabbix版本兼容性是必须优先考虑的一点客户端AGENT兼容 zabbix1.x到zabbix2.x的所有agent都兼容zabbix server2.4：如果你升级zabbix server，客户端是可以不做任何改变，除非你想使用agent的一些新特性。 Zabbix代理（p
unity 3d还是cocos2dx哪个适合游戏？ brotherlamp unity自学 unity教程 unity视频 unity资料 unity
unity 3d还是cocos2dx哪个适合游戏？问：unity 3d还是cocos2dx哪个适合游戏？答：首先目前来看unity视频教程因为是3d引擎，目前对2d支持并不完善，unity 3d 目前做2d普遍两种思路，一种是正交相机，3d画面2d视角，另一种是通过一些插件，动态创建mesh来绘制图形单元目前用的较多的是2d toolkit，ex2d，smooth moves，sm2，
百度笔试题：一个已经排序好的很大的数组，现在给它划分成m段，每段长度不定，段长最长为k，然后段内打乱顺序，请设计一个算法对其进行重新排序 bylijinnan java 算法面试百度招聘
import java.util.Arrays; /** * 最早是在陈利人老师的微博看到这道题： * #面试题#An array with n elements which is K most sorted，就是每个element的初始位置和它最终的排序后的位置的距离不超过常数K * 设计一个排序算法。It should be faster than O(n*lgn)。
获取checkbox复选框的值 chiangfai checkbox
<title>CheckBox</title> <script type = "text/javascript"> doGetVal: function doGetVal() { //var fruitName = document.getElementById("apple").value;//根据
MySQLdb用户指南 chenchao051 mysqldb
原网页被墙，放这里备用。 MySQLdb User's Guide Contents Introduction Installation _mysql MySQL C API translation MySQL C API function mapping Some _mysql examples MySQLdb
HIVE 窗口及分析函数 daizj hive 窗口函数分析函数
窗口函数应用场景：（1）用于分区排序（2）动态Group By （3）Top N （4）累计计算（5）层次查询一、分析函数用于等级、百分点、n分片等。函数说明 RANK() &nbs
PHP ZipArchive 实现压缩解压Zip文件 dcj3sjt126com PHP zip
PHP ZipArchive 是PHP自带的扩展类，可以轻松实现ZIP文件的压缩和解压，使用前首先要确保PHP ZIP 扩展已经开启，具体开启方法就不说了，不同的平台开启PHP扩增的方法网上都有，如有疑问欢迎交流。这里整理一下常用的示例供参考。一、解压缩zip文件 01 02 03 04 05 06 07 08 09 10 11
精彩英语贺词 dcj3sjt126com 英语
I'm always here 我会一直在这里支持你 &nb
基于Java注解的Spring的IoC功能 e200702084 java spring bean IOC Office
java模拟post请求 geeksun java
一般API接收客户端（比如网页、APP或其他应用服务）的请求，但在测试时需要模拟来自外界的请求，经探索，使用HttpComponentshttpClient可模拟Post提交请求。此处用HttpComponents的httpclient来完成使命。 import org.apache.http.HttpEntity ; import org.apache.http.HttpRespon
Swift语法之 ---- ?和!区别 hongtoushizi ?swift !
转载自： http://blog.sina.com.cn/s/blog_71715bf80102ux3v.html Swift语言使用var定义变量，但和别的语言不同，Swift里不会自动给变量赋初始值，也就是说变量不会有默认值，所以要求使用变量之前必须要对其初始化。如果在使用变量之前不进行初始化就会报错： var stringValue : String //
centos7安装jdk1.7 jisonami jdk centos
安装JDK1.7 步骤1、解压tar包在当前目录 [root@localhost usr]#tar -xzvf jdk-7u75-linux-x64.tar.gz 步骤2：配置环境变量在etc/profile文件下添加 export JAVA_HOME=/usr/java/jdk1.7.0_75 export CLASSPATH=/usr/java/jdk1.7.0_75/lib
数据源架构模式之数据映射器 home198979 PHP 架构数据映射器 datamapper
前面分别介绍了数据源架构模式之表数据入口、数据源架构模式之行和数据入口数据源架构模式之活动记录，相较于这三种数据源架构模式，数据映射器显得更加“高大上”。一、概念数据映射器（Data Mapper）：在保持对象和数据库（以及映射器本身）彼此独立的情况下，在二者之间移动数据的一个映射器层。概念永远都是抽象的，简单的说，数据映射器就是一个负责将数据映射到对象的类数据。 &nb
在Python中使用MYSQL pda158 mysql python
缘由　　近期在折腾一个小东西须要抓取网上的页面。然后进行解析。将结果放到数据库中。　　了解到 Python在这方面有优势，便选用之。　　由于我有台 server上面安装有 mysql，自然使用之。在进行数据库的这个操作过程中遇到了不少问题，这里记录一下，大家共勉。　　 python中mysql的调用　　百度之后能够通过MySQLdb进行数据库操作。
单例模式 hxl1988_0311 java 单例设计模式单件
package com.sosop.designpattern.singleton; /* * 单件模式：保证一个类必须只有一个实例，并提供全局的访问点 * * 所以单例模式必须有私有的构造器，没有私有构造器根本不用谈单件 * * 必须考虑到并发情况下创建了多个实例对象 * */ /** * 虽然有锁，但是只在第一次创建对象的时候加锁，并发时不会存在效率
27种迹象显示你应该辞掉程序员的工作 vipshichg 工作
1、你仍然在等待老板在2010年答应的要提拔你的暗示。 2、你的上级近10年没有开发过任何代码。 3、老板假装懂你说的这些技术，但实际上他完全不知道你在说什么。 4、你干完的项目6个月后才部署到现场服务器上。 5、时不时的，老板在检查你刚刚完成的工作时，要求按新想法重新开发。 6、而最终这个软件只有12个用户。 7、时间全浪费在办公室政治中，而不是用在开发好的软件上。 8、部署前5分钟才开始测试。

MyDLNote - Attention: [NLA系列] Asymmetric Non-local Neural Networks for Semantic Segmentation

Asymmetric Non-local Neural Networks for Semantic Segmentation

Abstract

Introduction

Related Work

Asymmetric Non-local Neural Network

Revisiting Non-local Block

Asymmetric Pyramid Non-local Block

Asymmetric Fusion Non-local Block

Network Architecture

Experiments

Datasets and Evaluation Metrics

Implementation Details

Comparisons with Other Methods

Conclusion

你可能感兴趣的:(deep,learning,semantic,segmentation,non-local)