How to Use Your Raspberry Pi to Detect Objects (TensorFlow + Camera Module)

preview_player
Показать описание
I will demonstrate how we can use our Raspberry Pi Model B+ to detect ANY object captured in the camera frames or video. Specifically we create a model that looks for instances of guitars within pictures and returns a probability that the image contains a guitar. We use TensorFlow, Docker, and the Macgyver API to string together these technologies and deploy them to our edge device. This video will walk you through each step so that you will be able to do the same.

#tensorflow #raspberrypi #computervision

Camera Module Used

Wayixia Batch Download Chrome Extension

Macgyver Tensor Flow Docker Container

Рекомендации по теме
Комментарии
Автор

Nice job! I have a working mobilenet model on the pi and would like to use the camera input for real time image classification, rather than just classifying one image at a time. Can you elaborate on the script you used for this?

styfrsy
Автор

It can also accepted as a decent Docker tutorial?

mcan
Автор

You think its possible storing the TF model locally on the Raspberry Pi? Would be great to have real time object detection without internet or a server. Great video, regardless, subbed!

revhappyrsk
Автор

We are doing waste classification identifier...so we have to open the particular(recyclable, organic, harmful) waste box...when the waste is classified..so if we do cloud based system ..how can we open the box through raspberry pi

rahulnpr
Автор

Hello, I'm a student . I want to use rasbarrypi to identify the traffic light color, recognize the color, stop or walk with my voice. Please give us some advice and direction. Thank you .

沈郁晉-my
Автор

Hey, How to increase pi video processing, because it got struck when detecting object.

salimattari
Автор

Hey there, your vid is great. Since I'm new to RPi, i have one question
1. Is the software u use to code Python IDE? Or terminal?
Thanks for the help!

isaacchook
Автор

is there a way in which i can develop system to identify defects say any cracks, stains or holes in the detected object by camera?

sanyam.jain
Автор

Is that fast? should i using Movidious to accelerate it ?

jackyli
Автор

wow, that's great! did you using tensorflow and openCV ?

jackyli
Автор

Can you please tell what "algorithm" does tensorflow object detection api use for classification?

shayanbarua
Автор

The macgyvertechnology/tensorflow container exit immediately, any idea what goes wrong?

squall
Автор

I keep getting an exit message in the status for some reason, I ran the command "docker run -it -d Any idea?
CONTAINER ID IMAGE COMMAND CREATED STATUS PORTS NAMES
298fdb2fc819 macgyvertechnology/tensorflow "/bin/bash" 12 seconds ago Exited (1) 11 seconds ago pedantic_gauss

drrufusjones
Автор

I keep receiving the error after sorting the $docker ps -a "Exited (1) About a minute ago" Please has anyone found the solution for this?

azubuke
Автор

Can we use camera module 5MP raspberry to detect object?

bertolomeouzbezozdiaz
Автор

Why the demo isn't working now ???

yashshah
Автор

Sir can you upload the node.js script??

shayanbarua
Автор

Why did you implement your classifier in a docker container, instead of .... outside the container in your pc ?!
*noob alert

staynjohnson
Автор

Job offer regarding the same contents from this video. If you are interested and confident then pm me.
Thanks

nickarchin
Автор

My word this technology fad sucks balls.. Fucking Google junk...

Diggnuts