论文阅读：Learning Human-Object Interactions by Graph Parsing Neural Networks

最新推荐文章于 2025-07-17 19:47:42 发布

原创

最新推荐文章于 2025-07-17 19:47:42 发布 · 3.3k 阅读

7 ·

CC 4.0 BY-SA版权

文章标签：

#scene understand

GPNN(ECCV2018)

文章
用来检测human-object interaction。Paper说提出GPNN的目的是为了在HOI任务中显式地对空间、时间（视频）以及human-object关系进行利用。
这里写图片描述
complete HOI指的是所有的node之间都有link，node包括human和object，假设node特征为d_V维，edge的特征为d_E维（从两个box的union box提取得到的），且complete HOI一共有N个node，那么特征矩阵F则为NxNx(2d_V+d_E)，也就是每两个node及其之间的边都要按顺序组合一次。然后用1x1的卷积计算出一个邻接矩阵。邻接矩阵可以用来表示两个node的某种相关性

上面的公式为每个node的状态h的更新过程，A是邻接矩阵，M是计算message的函数，U是利用当前状态h和新的信息m计算更新状态h的函数。然后这样不断地迭代：

计算特征矩阵F –> 计算邻接矩阵A –> 计算message –> 更新node隐藏状态 –> 计算特征矩阵F…

迭代一定次数后，graph收敛？得到每个node的一个综合了context的隐藏状态，然后使用一个readout函数，得到每个node的输出：

最低0.47元/天解锁文章

确定要放弃本次机会？

福利倒计时

: :

立减 ¥

普通VIP年卡可用

立即使用

Kivee123

关注关注

0
点赞
踩
7

收藏

觉得还不错? 一键收藏
0
评论
分享

复制链接

分享到 QQ

分享到新浪微博

扫一扫
举报

举报

Graph Neural Networks: A Review of Methods and Applications（图神经网络：方法与应用综述）

weixin_41697507的博客

07-06

5887

Graph Neural Networks: A Review of Methods and Applications Jie Zhou , Ganqu Cui , Zhengyan Zhang , Cheng Yang, Zhiyuan Liu, Lifeng Wang, Changcheng Li, Maosong Sun Abstract—Lots of learning tasks req...

Learning Human-Object Interactions by Graph Parsing Neural Networks阅读笔记

qq_46805191的博客

03-17

662

前言这是一篇2018年发表在ECCV上得paper，论文原文点这里。笔记说实话刚看到这篇文章还是有点唬人的，文中作者提出了一个网络模型叫GPNN，图解析神经网络，我以为是跟GNN有关系，花了一些时间去看了近些年GCN的一些发展回头又读这篇文章的时候把GCN的思想带进去发现怎么都读不懂，后来才发现这篇文章仅仅是用了图的“结构”，后边的网络仍然是正常的Conv2D卷积。作者介绍，其提出了GPNN，一个可以结合结构知识和深度神经网络的学习能力的端到端的模型框架。对于给定的情景（一张图片），GPN

参与评论您还未登录，请先登录后发表或查看评论

Learning Human-Object Interactions by Graph Parsing Neural Networks代码学习（Ubuntu+pycharm）

qq_40250617的博客

05-19

1921

１．class GPNN_HICO(torch.nn.Module)：　类　　torch.nn.Module这个类内部有多达４８个函数，是ｐｙｔｏｒｃｈ中所有神经网络模型的基类；GPNN_HICO继承了torch.nn.Module这个类，成为了torch.nn.Module的子类，２．def __init__(self, model_args):　方法　　ｓｅｌｆ是定义每个类的方法 ...

ECCV 2018| Learning Human-Object Interactions by Graph Parsing Neural Networks阅读笔记

weixin_59340037的博客

04-20

1126

本文主要是为了解决图像和视频中的HOI任务，引入了图解析神经网络（GPNN），是一个利用结构知识的端到端的可分辨结构。给出一个场景，GPNN能够推断出一个解析图，并且包含（1）使用邻接矩阵表示的图结构（2）节点标签。当消息传递到网络中是，GPNN就会迭代计算出邻接矩阵与节点标签。在V-COCO、HICO-DET、CAD-120上评估该方法都取得了较好的效果。论文地址：http://cn.arxiv.org/abs/1808.07962 代码地址：https://github.com/SiyuanQi/gp

Detecting and Recognizing Human-Object Interactions

10-12

要理解视觉世界，机器不仅必须识别单个目标，还必须识别它们是如何交互的。本文提出了一种新的模型，挑战日常照片中检测⟨人类、动词、目标⟩三元组的任务。

HOTR: End-to-End Human-Object Interaction Detection with Transformers

weixin_43889476的博客

06-01

699

使用Transformer实现HOI检测,附带自己画的框架图,比较清晰的带你理解HOI检测的流程.

论文笔记之Learning Human-Object Interaction Detection using Interaction Points

加油

07-25

2871

直接预测人-物的交互点和交互向量（用于分组），再与人/物检测结果关联得到最终结果。 CVPR2020接收论文地址：https://arxiv.org/abs/2003.14023 1. 摘要理解人与物体之间的相互作用是视觉分类的基本问题之一，也是实现详细场景理解的重要步骤。人与物体之间的交互（HOI）检测力求既定位人与物体，又确定它们之间的复杂交互。大多数现有的HOI检测方法都是以实例为中心的，其中基于外观特征和粗糙的空间信息来预测所有可能的人-物体之间的交互。作者认为，仅外观特征不足以捕获复杂.

人物交互（human object interaction）论文汇总-2019年

加油

11-08

3457

1. Relation Parsing Neural Network for Human-Object Interaction Detection 1.1 总述提出一种关系解析神经网络RPNN，由两部分组成：物体-身体部位图和人体-身体部位图，前者捕获身体部位与周围物体的关系，后者推断人体与身体部位的关系，并组合身体部位上下文信息以预测动作。这两个图由动作传递机制关联。总的来说，网络基于人的特征预测动作发生的位置（参与交互物体的位置）的密度；基于物体特征预测每个动作的物体与人之间交互的概率。 1.2 网

READING NOTE: Learning to Detect Human-Object Interactions

joshua_1988

02-28

1249

TITLE: Learning to Detect Human-Object Interactions

论文阅读

parker_1的博客

08-12

224

参考： https://blog.youkuaiyun.com/zhanshen112/article/details/81745726 https://blog.youkuaiyun.com/zhanshen112/article/details/81745726

CVPR2018 Detecting and Recognizing Human-Object Interactions阅读笔记

yang_daxia的博客

10-11

417

1.Introduction 本文不仅检测object, 还把object分为human和物体, 考虑任何物体的关联关系. 目的是学习一个三元组关系<human, action, object> 2.method 作者在fast rcnn基础上扩展来个分支, 获得三元组的score, 该score由4部分组成: Sh,oa=sh⋅so⋅sha⋅gh,oa(1) S_{h, o}^...

图像理解--Detecting and Recognizing Human-Object Interactions

AI小作坊的博客

06-05

5437

Detecting and Recognizing Human-Object Interactions https://arxiv.org/abs/1704.07333大牛们已经从图像的检测分割向图像理解的研究方向过度了！本文主要关注图像中的人和物体的关系检测和识别，这种关系可以用一个三元素

Human-Object Interaction (HOI) synthesis

轻舟已过万重山

07-17

601

Human-Object Interaction (HOI) synthesis refers to the generation of realistic interactions between humans and objects in virtual or physical environments. Unlike HOI recognition (which focuses on detecting existing interactions), HOI synthesis aims to cre

论文阅读：Interact as You Intend: Intention-Driven Human-Object Interaction Detection

Kivee的博客

10-09

1436

iHOI（submitted to IEEE Trans. on Multimedia）文章本文focus在检测social scene images上的human-object interactions（HOIs）。本文的一个出发点是human在与object进行interaction的时候，会根据他们的intention调整注意力和移动身体，因此本文的就在HOI检测中加入了human ...

论文笔记：Transferable Interactiveness Knowledge for Human-Object Interaction Detection(CVPR2019)

王小石的博客

05-11

370

核心思想：学习Interactiveness Knowledge来判断是否存在人与物体的交互。用图来表示图片中的实例和关系，实例是节点，关系是边，如果采用穷举法配对，将会产生大量的边，其中大部分没有交互关系的边都应该被舍弃。思路：任何HOI数据集都可以转换成二分类标签，用来学习Interactiveness Knowledg。先用Interactiveness Knowledge进行Non-Interaction抑制，得到稀疏图，再进行交互关系的分类。具体： 1.包括三部分：特征表示R、交互判断P、交互

论文阅读（一）

lwyeluo的专栏

01-04

1986

所有论文笔记请戳：http://blog.youkuaiyun.com/lwyeluo/article/details/54017718 Trusted Geolocation-Aware Data Placement infrastructure clouds 2015-12-09 摘要云中数据的地理定位成为亟待解决的问题，尤其是在不同行政区域法规与数据所有者需求不兼容的情况下。本

Human-AI Interactions

最新发布

08-12

### Human-AI Interaction Design Principles and Best Practices Designing effective interactions between humans and AI systems involves a set of principles and best practices aimed at enhancing usability, trust, and efficiency. These principles ensure that AI systems are not only functional but also intuitive and accessible to a wide range of users. One of the key aspects of human-AI interaction design is ensuring that the system is accessible to all users, including those with disabilities. This aligns with the concepts of universal design, inclusive design, and accessible design, which emphasize increasing the accessibility of interactive systems such as websites, browsers, tools, and various digital products. These design philosophies share a common goal of making technology usable by everyone, regardless of their abilities or disabilities [^1]. When designing for human-AI interaction, it is essential to consider the following principles: - **Transparency**: Users should understand how the AI system works and why it makes certain decisions. This helps build trust and allows users to predict the system's behavior. - **Control and Flexibility**: Users should have control over the AI system and be able to customize it according to their needs. This includes providing options for adjusting settings and preferences. - **Feedback and Error Handling**: The system should provide clear feedback and handle errors gracefully. Users should be informed when something goes wrong and given guidance on how to correct it. - **Consistency**: Consistent design elements and behaviors help users learn and use the system more effectively. Consistency should be maintained across different parts of the system and over time. - **Simplicity and Intuitiveness**: The interface should be simple and easy to use. Complex features should be hidden until needed, and the system should be designed to be as intuitive as possible. Implementing these principles requires a deep understanding of both the technical capabilities of AI systems and the needs of the users. For instance, when developing complex LabVIEW applications using the Actor Framework, it is crucial to evaluate the advantages and challenges associated with building large, concurrent, and scalable systems. This involves researching the principles, design patterns, and implementation details of the Actor Framework to ensure that the resulting applications meet high standards of performance and reliability [^2]. Furthermore, considering the competitive landscape in the test and measurement and automation fields, it is important to stay ahead by incorporating innovative design practices and leveraging the latest technologies to enhance user experience and system performance [^3]. ### Code Example Here is a simple example of how to implement a basic feedback mechanism in a Python application: ```python def get_user_feedback(): feedback = input("Please provide your feedback: ") print(f"Thank you for your feedback: {feedback}") get_user_feedback() ``` This code snippet demonstrates a function that collects user feedback and acknowledges it, which is a fundamental aspect of error handling and user engagement in human-AI interaction design. ###