Roboflow 6 Minute Intro | Build a Coin Counter with Computer Vision

Create a dataset and train a model that can see individual coins, sum their value, and run in real-time (or via API). This step-by-step video demonstrates an overview of how Roboflow accelerates building vision capabilities, including from image collection to annotation and training to deployment and improvement. For a deeper dive, reference our docs: docs.roboflow.com
Project available on Universe: universe.robof...

[@nazardudyk2611]

Simply amazing! It looks so shockingly simple. I mean, all the work behind this is incredibly intricate. Congrats! I’ll definitely try it 100%.

[@danielisflying]

Amazing intro to Roboflow and Object detection. Very intuitive platform for dataset creation.

[@Roboflow]

Thank you!

[@hongbo-wei]

Amazing, Reboflow makes utilizing computer vision so easy! Great job! Much appreciate it!

[@littlebrit]

What do you use to select coins? Is there a Javascript library for that? Sorry, just learning coding.

[@tryingtobeproductive]

Wow. this really needs to be more well known.

[@dlrlanders892]

Hello, thank you for the video - really helpful! At @2:56, you said you used your "money counter" model which you already created. How did you create this? Is this from a previous project? I am trying to use model assisted labelling with my images because I have over 15,000 images to get through. How do I create a model to help me label my images? thanks again!

[@Roboflow]

Hello! You are able to use any public model to power model assisted labeling: blog.roboflow.com/launch-universe-model-checkpoint/ With your 15,000 images, you'd then label the data and train a model. Once you have a working model, you can use it label more images. So, you could label 200 images then train a model which you will use to automate labeling the other 14,000+ images. This video guide shows how to train a model in depth: kzbin.info/www/bejne/hV69gauKpcdlj8k

[@chaztikov]

Great intro.

[@oxydol3456]

Nice tutorial. The training status has been stuck at "Training machine starting..." for hours. The number of custom images of mine is 10. Each size is around 2mb. jpeg. Could overlapping bounding boxes be the root?

[@vishnusudheer6049]

Hi i have a question regarding your drone detection model. How can i use your drone detection model hosted api code in python so that it detecta drones on a webcam feed. I tried doing it on pycharm but its giving me errors regarding cv installation etc. Can you help me

[@wiamrabhi4420]

When you start training in roboflow from public chackpoint what model (Yolo, SSD...) will be training? Can we change it?

[@kees5944]

Amazing thanks

[@mrajith303]

Wow, I was recently working on the same project, found it little difficult with the shining photos, even with gaussian blur and dilation. This is easier to use. will try it out

[@Roboflow]

Great to hear. The project in this video is open source and on Universe: universe.roboflow.com/corn-counting/money-counter/health

[@jaybryanodero8110]

Hello sir just wanna ask if I can use this methods to identify citrus fruits that are similar color

[@fifthperson9777]

Nice, maybe in another video demonstrate how we can deploy them on android, offline

[@kevj1605]

How do you provide a video input feed to that?

[@dusanigic2608]

I don't understand how do you got 17fps with webcame I have only 1 fps. Please some help

[@erickgremeaux7846]

Hello, i can't export m'y dataset, il dont have yolov.... choice... What is thé problème

[@zhuwei5926]

If the object is partially overlapping, how good is the recognition?

[@BiggestRetardInTheWorld]

Try polygon boxing

[@gonzaloillansalvador5858]

What is the cost of using that model?

[@zenbauhaus1345]

Unbelievable