当前位置:网站首页>My brother created his own AI anti procrastination system, and he was "blinded" when playing with his mobile phone | reddit was hot
My brother created his own AI anti procrastination system, and he was "blinded" when playing with his mobile phone | reddit was hot
2022-07-26 15:49:00 【QbitAl】
Alex From the Aofei temple
qubits | official account QbitAI
You have been Deadline Fear of domination ?

In order to overcome Procrastination , A little brother to himself “ It's hard ”.
He wrote a set of using Object detection API The algorithm of , To supervise their focus on learning .
Once he starts playing with his cell phone , The punishment system will start ——
The two flashlights in front of me will “ Light blind ” His eyes .

meanwhile , The stereo will also make a harsh sound , Force him back to his current task .
Brother shared this self created anti delay system to YouTube and Reddit, It turns out that Reddit It's on fire , So far 3.3k degree of heat .

Many netizens exclaimed :Amazing!
A place “ Late procrastination ” Express , If this punishment system can spray pepper water , It should be very effective for him .
Others even put forward More exciting Punishment measures :
Every time I lose my mind , Just deduct money from your credit card .

however , It was also suggested that :
If you put the pencil on the back of the phone , The system will recognize which ?

How does this anti delay system work ?
Brother used it directly TensorFlow Upper object detection API, To build a target detection framework , Used to identify objects within the range of the camera .
Through this object detection API, The computer can recognize multiple objects at the same time .
Brother set a productivity score, Different objects will have their corresponding scores .

Computer according to productivity score To identify each object .


For this system , Little brother writes Algorithm , Set the phone as the primary “ Distractions ”. in other words , When the computer recognizes the mobile phone , Will judge that people are distracted .
And when a pencil is detected , The system thinks he is doing problems , Will not trigger punishment :

He demonstrated the effect of this system in the video .
When he is focused ( There are no objects such as mobile phones in the camera's field of view ), The computer will display in real time “On Task”.

And when he picked up his cell phone , The computer will display “Off Task”.

below , How to make the computer and punishment equipment linked ?
Brother created a database , Whenever the computer detects that people are distracted , Will send some data to the database , The database will update the server in real time according to this value .
The value in this server , The switch that will control the punishment device .

The punishment equipment here , Include 2 A flashlight and 1 A stereo .
When the system detects distractions , The flashlight will automatically emit strong light , At the same time, the stereo will also make a harsh sound .

Outside “ Physical stimulation ” Next , People in front of the computer screen will have to focus again .
It can be said that this is a “ Ancient early ” The target detection technology of is used again .
See here , You think this little brother's method is right “ Treatment ” Is procrastination effective enough ?
Or do you have any urge due Great idea ? Feel free to leave a comment in the comments section ~
Code address :
https://github.com/ayushpai/Object-Detector
Reference link :
[1]https://www.reddit.com/r/MachineLearning/comments/vkxsf2/i_made_a_robot_that_punishes_me_if_it_detects/
[2]https://www.youtube.com/watch?v=YPSazrEqlxo
边栏推荐
- QT is the most basic layout, creating a window interface
- ES6 advanced - query commodity cases
- 777. Exchange adjacent characters in LR string
- Quanzhi a40i industrial core board, 100% domestic 4-core arm cortex-a7, supports "dual screen abnormal display" [display interface capability, preferred scheme for industrial HMI]
- [expdp export data] expdp exports a table with 23 rows of records and no lob field. It takes 48 minutes. Please help us have a look
- R language uses LM function to build a multiple regression model with interactive terms, and uses step function to build a stepwise regression model to screen the best subset of predictive variables (
- QCF for deep packet inspection paper summary
- [leetcode daily question] - 268. Missing numbers
- Creation and traversal of binary tree
- Tool skill learning (I): pre skills -makfile, make,.Mk
猜你喜欢

04 callable and common auxiliary classes

kalibr标定realsenseD435i --多相机标定
原来卡布奇诺信息安全协会是干这个的呀,一起来看看吧。

TI C6000 TMS320C6678 DSP+ Zynq-7045的PS + PL异构多核案例开发手册(3)

Data preprocessing of data mining

Tutorial (7.0) 05. Issue forticlient * forticlient EMS * Fortinet network security expert NSE 5 through forticlient EMS

【DSCTF2022】pwn补题记录

A comprehensive review of image enhancement technology in deep learning

FOC电机控制基础

Zynq PS + PL heterogeneous multicore Case Development Manual of Ti C6000 tms320c6678 DSP + zynq-7045 (1)
随机推荐
中金财富炒股安全吗 手续费最便宜的证券公司
【五分钟Paper】基于参数化动作空间的强化学习
University rankings in Beijing
Encryption model
Driver development environment
FTP协议
Data preprocessing of data mining
Sword finger offer II 009. subarray with product less than k
OSPF comprehensive experiment
R language uses LM function to build multiple linear regression model, writes regression equation according to model coefficient, and uses fitted function to calculate y value (response value) vector
组件化开发基本规范、localStorage 和 sessionStorage、对象数据转基本值、原型链使用
13年资深开发者分享一年学习Rust经历:从必备书目到代码练习一网打尽
Can't you see the withdrawal? Three steps to prevent withdrawal on wechat.
VP video structured framework
TI C6000 TMS320C6678 DSP+ Zynq-7045的ZYNQ PS + PL异构多核案例开发手册(1)
Digital warehouse: iqiyi digital warehouse platform construction practice
What is a virtual camera
[leetcode daily question] - 121. The best time to buy and sell stocks
js 对数组操作的 API 总结
Tool skill learning (I): pre skills -makfile, make,.Mk