GBIM(Gesture-Based Interaction map)

Overview

GBIM

Python 3.6 PaddleX License

手势交互地图 GBIM(Gesture-Based Interaction map),基于视觉深度神经网络的交互地图,通过电脑摄像头观察使用者的手势变化,进而控制地图进行简单的交互。网络使用PaddleX提供的轻量级模型PPYOLO Tiny以及MobileNet V3 small,使得整个模型大小约10MB左右,即使在CPU下也能快速定位和识别手势。

手势

手势 交互 手势 交互 手势 交互
向上滑动 向左滑动 地图放大
手势 交互 手势 交互 手势 交互
向下滑动 向右滑动 地图缩小

进度安排

基础

  • 确认用于交互的手势。
  • 使用det_acq.py采集一些电脑摄像头拍摄的人手姿势数据。
  • 数据标注,训练手的目标检测模型
  • 捕获目标手,使用clas_acq.py获取手部图像进行标注,并用于训练手势分类模型。
  • 交互手势的检测与识别组合验证。
  • 打开百度地图网页版,进行模拟按键交互。
  • 组合功能,验证基本功能。

进阶

  • 将图像分类改为序列图像分类,提高手势识别的流畅度和准确度。
  • 重新采集和标注数据,调参训练模型。
  • 搭建可用于参数调节的地图。
  • 界面整合,整理及美化。

数据集 & 模型

手势检测

  • 数据集使用来自联想小新笔记本摄像头采集的数据,使用labelImg标注为VOC格式,共1011张。该数据集场景、环境和人物单一,仅作为测试使用,不提供数据集下载。数据组织参考PaddelX下的PascalVOC数据组织方式。
  • 模型使用超轻量级PPYOLO Tiny,模型大小小于4MB,随便训练了100轮后保留best_model作为测试模型,由于数据集和未调参训练的原因,当前默认识别效果较差

手势分类

  • 数据集使用来自联想小新笔记本摄像头采集的数据,通过手势检测模型提出出手图像,人工分为7类,分别为6种交互手势以及“其他”,共1102张。该数据集数量较少,手型及手势单一,仅作为测试使用,不提供数据集下载。数据组织形式如下:
dataset
	├-- Images
	|     ├-- up
	┆     ┆    └-- xxx.jpg
	|     └-- other
	┆          └-- xxx.jpg
	├-- labels.txt
	├-- train_list.txt
	└-- val_list.txt
  • 模型使用超轻量级MobileNet V3 small,模型大小小于7MB,由于数据量很小,随便训练了20轮后保留best_model作为测试模型,当前识别分类效果较差

模型文件上传使用LFS,下拉时注意需要安装LFS,参考LFS文档。后续将重新采集和标注更加多样的大量数据集,并采用更好的调参方法获得更加准确的识别模型

演示

手势识别

地图交互

*未显示Capture界面

使用

  1. 克隆当前项目到本地,按照requirements.txt安装所依赖的包opencv、paddlex以及pynput。PaddleX对应请安装最新版的PaddlePaddle,由于模型轻量,CPU版本足矣,参考下面代码,细节参考官方网站
python -m pip install paddlepaddle -i https://mirror.baidu.com/pypi/simple
  1. 进入demo.py,将浏览器路径修改为自己使用的浏览器路径:
web_path = '"D:/Twinkstar/Twinkstar Browser/twinkstar.exe"'  # 自己的浏览器路径
  1. 运行demo.py启动程序:
cd GBIM
python demo.py

常见问题及解决

  1. Q: 拉项目时卡住不动

    A:首先确认按照文档安装LFS。如果已经安装那极大可能是网络问题,可以等待一段时间,或先跳过LFS文件,再单独拉取,参考下面git代码:

    // 开启跳过无法clone的LFS文件
    git lfs install --skip-smudge 
    // clone当前项目
    git clone "current project" 
    // 进入当前项目,单独拉取LFS文件
    cd "current project" 
    git lfs pull 
    // 恢复LFS设置
    git lfs install --force
  2. Q:按q或者手势交互无效

    A:请注意当前鼠标点击的焦点,焦点在Capture,则接受q退出;焦点在浏览器,则交互结果将驱动浏览器中的地图进行变换。

  3. Q:安装PaddleX时报错,关于MV C++

    A:若在Windows下安装coco tool时报错,则可能缺少Microsoft Visual C++,可在微软官方下载网页进行下载安装后重启,即可解决。

  4. Q:运行未报错,但没有保存数据到本地

    A:请检查路径是否有中文,cv2.imwrite保存图像时不能有中文路径。

参考

  1. 玩腻了小游戏?Paddle手势识别玩转游戏玩出新花样!
  2. https://github.com/PaddlePaddle/PaddleX

交流与反馈

Email:[email protected]

Code for MarioNette: Self-Supervised Sprite Learning, in NeurIPS 2021

MarioNette | Webpage | Paper | Video MarioNette: Self-Supervised Sprite Learning Dmitriy Smirnov, Michaël Gharbi, Matthew Fisher, Vitor Guizilini, Ale

Dima Smirnov 28 Nov 18, 2022
Official implementation for the paper "Attentive Prototypes for Source-free Unsupervised Domain Adaptive 3D Object Detection"

Attentive Prototypes for Source-free Unsupervised Domain Adaptive 3D Object Detection PyTorch code release of the paper "Attentive Prototypes for Sour

Deepti Hegde 23 Oct 17, 2022
Multi-Stage Progressive Image Restoration

Multi-Stage Progressive Image Restoration Syed Waqas Zamir, Aditya Arora, Salman Khan, Munawar Hayat, Fahad Shahbaz Khan, Ming-Hsuan Yang, and Ling Sh

Syed Waqas Zamir 859 Dec 22, 2022
Info and sample codes for "NTU RGB+D Action Recognition Dataset"

"NTU RGB+D" Action Recognition Dataset "NTU RGB+D 120" Action Recognition Dataset "NTU RGB+D" is a large-scale dataset for human action recognition. I

Amir Shahroudy 578 Dec 30, 2022
Multi-angle c(q)uestion answering

Macaw Introduction Macaw (Multi-angle c(q)uestion answering) is a ready-to-use model capable of general question answering, showing robustness outside

AI2 430 Jan 04, 2023
CLIPImageClassifier wraps clip image model from transformers

CLIPImageClassifier CLIPImageClassifier wraps clip image model from transformers. CLIPImageClassifier is initialized with the argument classes, these

Jina AI 6 Sep 12, 2022
Research Artifact of USENIX Security 2022 Paper: Automated Side Channel Analysis of Media Software with Manifold Learning

Automated Side Channel Analysis of Media Software with Manifold Learning Official implementation of USENIX Security 2022 paper: Automated Side Channel

Yuanyuan Yuan 175 Jan 07, 2023
Yoga - Yoga asana classifier for python

Yoga Asana Classifier Description Hi welcome to my new deep learning project "Yo

Programminghut 35 Dec 12, 2022
Multi-Task Deep Neural Networks for Natural Language Understanding

New Release We released Adversarial training for both LM pre-training/finetuning and f-divergence. Large-scale Adversarial training for LMs: ALUM code

Xiaodong 2.1k Dec 30, 2022
(AAAI2022) Style Mixing and Patchwise Prototypical Matching for One-Shot Unsupervised Domain Adaptive Semantic Segmentation

SM-PPM This is a Pytorch implementation of our paper "Style Mixing and Patchwise Prototypical Matching for One-Shot Unsupervised Domain Adaptive Seman

W-zx-Y 10 Dec 07, 2022
FinGAT: A Financial Graph Attention Networkto Recommend Top-K Profitable Stocks

FinGAT: A Financial Graph Attention Networkto Recommend Top-K Profitable Stocks This is our implementation for the paper: FinGAT: A Financial Graph At

Yu-Che Tsai 64 Dec 13, 2022
利用Tensorflow实现基于CNN的中文短文本分类

Text Classification with CNN 使用卷积神经网络进行中文文本分类 CNN做句子分类的论文可以参看: Convolutional Neural Networks for Sentence Classification 还可以去读dennybritz大牛的博客:Implemen

Jeremiah 4 Nov 08, 2022
Tensorflow implementation of ID-Unet: Iterative Soft and Hard Deformation for View Synthesis.

ID-Unet: Iterative-view-synthesis(CVPR2021 Oral) Tensorflow implementation of ID-Unet: Iterative Soft and Hard Deformation for View Synthesis. Overvie

17 Aug 23, 2022
A collection of 100 Deep Learning images and visualizations

A collection of Deep Learning images and visualizations. The project has been developed by the AI Summer team and currently contains almost 100 images.

AI Summer 65 Sep 12, 2022
A Collection of LiDAR-Camera-Calibration Papers, Toolboxes and Notes

A Collection of LiDAR-Camera-Calibration Papers, Toolboxes and Notes

443 Jan 06, 2023
Structure-Preserving Deraining with Residue Channel Prior Guidance (ICCV2021)

SPDNet Structure-Preserving Deraining with Residue Channel Prior Guidance (ICCV2021) Requirements Linux Platform NVIDIA GPU + CUDA CuDNN PyTorch == 0.

41 Dec 12, 2022
Single/multi view image(s) to voxel reconstruction using a recurrent neural network

3D-R2N2: 3D Recurrent Reconstruction Neural Network This repository contains the source codes for the paper Choy et al., 3D-R2N2: A Unified Approach f

Chris Choy 1.2k Dec 27, 2022
Free like Freedom

This is all very much a work in progress! More to come! ( We're working on it though! Stay tuned!) Installation Open an Anaconda Prompt (in Windows, o

2.3k Jan 04, 2023
🛠️ SLAMcore SLAM Utilities

slamcore_utils Description This repo contains the slamcore-setup-dataset script. It can be used for installing a sample dataset for offline testing an

SLAMcore 7 Aug 04, 2022