<p align="center">
<h1 align="center">QuickDraw - AirGesture</h1>
</p>
[![GitHub stars](https://img.shields.io/github/stars/uvipen/QuickDraw-AirGesture-tensorflow)](https://github.com/uvipen/QuickDraw-AirGesture-tensorflow/stargazers)
[![GitHub forks](https://img.shields.io/github/forks/uvipen/QuickDraw-AirGesture-tensorflow?color=orange)](https://github.com/uvipen/QuickDraw-AirGesture-tensorflow/network)
[![GitHub license](https://img.shields.io/github/license/uvipen/QuickDraw-AirGesture-tensorflow)](https://github.com/uvipen/QuickDraw-AirGesture-tensorflow/blob/master/LICENSE)
## Introduction
Here is my python source code for QuickDraw - an online game developed by google, combined with AirGesture - a simple gesture recognition application. By using my code, you could:
* **Run an app which you could draw in front of a camera with your hand (If you use laptop, your webcam will be used by default)**
* **Run an app which you could draw on a canvas**
## Camera app
In order to use this application, you only need to use your hand to draw in front of a camera/webcam. The middle point of your hand will be detected and highlighted by a red dot. When you are ready for drawing, you need to press **space** button to start drawing. When you want to stop drawing, press **space** button again.
Below is the demo by running the sript **camera_app.py**:
<p align="center">
<img src="demo/quickdraw_airgesture.gif" width=800><br/>
<i>Camera app demo</i>
</p>
## Drawing app
The script and demo will be released soon
## Categories:
The table below shows 18 categories my model used:
| | | | |
|-----------|:-----------:|:-----------:|:-----------:|
| apple | book | bowtie | candle |
| cloud | cup | door | envelope |
|eyeglasses | hammer | hat | ice cream |
| leaf | scissors | star | t-shirt |
| pants | tree | | |
## Trained models
You could find my trained model at **data/trained_models/**
## Docker
For being convenient, I provide Dockerfile which could be used for running training phase as well as launching application
Assume that docker image's name is qd_ag. You already clone this repository and cd into it.
Build:
`sudo docker build --network=host -t qd_ag .`
Run:
If you want to launch the application, first you need to run `xhost +` to turn off access control (if you only want to run the training, you could skip this step). Then you run:
`sudo docker run --gpus all -it --rm --volume="path/to/your/data:/workspace/code/data -e DISPLAY=$DISPLAY --env="QT_X11_NO_MITSHM=1" -v /tmp/.X11-unix:/tmp/.X11-unix --device=/dev/video0:/dev/video0 qd_ag`
Inside docker container, you could run **train.py** or **camera_app.py** scripts for training or launching app respectively. By default, the **camera_app.py** script will automatically generate a video capturing what you have done during the session, at **data/output.mp4**
## Experiments:
For each class, I split the data to training and test sets with ratio of 8:2. The training/test loss/accuracy curves for the experiment are shown below:
<img src="demo/loss_accuracy_curves.png" width="800">
没有合适的资源?快使用搜索试试~ 我知道了~
QuickDraw的实现-由Google开发的在线游戏,结合AirGesture-一个简单的手势识别应用程序_Pytho.zip
共32个文件
png:19个
py:5个
pb:2个
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 154 浏览量
2023-05-01
00:44:31
上传
评论
收藏 99.38MB ZIP 举报
温馨提示
QuickDraw的实现-由Google开发的在线游戏,结合AirGesture-一个简单的手势识别应用程序_Pytho.zip
资源推荐
资源详情
资源评论
收起资源包目录
QuickDraw的实现-由Google开发的在线游戏,结合AirGesture-一个简单的手势识别应用程序_Pytho.zip (32个子文件)
QuickDraw-AirGesture-tensorflow-main
drawing.py 4KB
src
utils.py 1KB
model.py 853B
data
pretrained_model.pb 17.92MB
trained_models
saved_model.pb 175KB
variables
variables.index 3KB
variables.data-00000-of-00001 7.39MB
LICENSE 1KB
demo
loss_accuracy_curves.png 92KB
quickdraw_airgesture.gif 76.23MB
camera_app.py 7KB
Dockerfile 345B
images
candle.png 6KB
pants.png 30KB
tree.png 6KB
hammer.png 5KB
cloud.png 4KB
envelope.png 3KB
eyeglasses.png 4KB
hat.png 8KB
door.png 4KB
apple.png 8KB
scissors.png 9KB
t-shirt.png 7KB
book.png 8KB
bowtie.png 8KB
cup.png 7KB
leaf.png 8KB
star.png 6KB
ice cream.png 6KB
train.py 4KB
README.md 3KB
共 32 条
- 1
资源评论
快撑死的鱼
- 粉丝: 1w+
- 资源: 9154
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
最新资源
- tiamo软件教程.doc
- nvm管理Nodejs多版本工具
- Unit1Howcanwebecomegoodlearners知识点整理(良心出品必属精品).doc
- 基于JSP在线维他茶饮销售网站平台源码.zip
- web学习笔记.doc
- 基于pytorch+Unet进行MRI肝脏图像分割源码+数据集+模型.zip
- 在Android Studio中开发一个Android App项目步骤
- 基于yolov8实现进行物体跟踪源码.zip
- Java多线程学习Java多线程学习Java多线程学习Java多线程学习.txt
- 算法数据结构-动态规划算法(Dynamic Programming)超详细总结加应用案例讲解.txt
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功