Pixy (CMUcam5) Smart Vision Sensor - Object Tracking Camera for Arduino, Raspberry Pi, BeagleBone Black
Fulfillment by Amazon (FBA) is a service we offer sellers that lets them store their products in Amazon's fulfillment centers, and we directly pack, ship, and provide customer service for these products. Something we hope you'll especially enjoy: FBA items qualify for FREE Shipping and .
If you're a seller, Fulfillment by Amazon can help you increase your sales. We invite you to learn more about Fulfillment by Amazon .
- Enter your model number to make sure this fits.
- Robot vision made easy - press the button to teach Pixy an object.
- Simplify your programming - receive just the objects you're interested in.
- Use whatever controller you want - includes software libraries for Arduino, Raspberry Pi, and BeagleBone Black.
- Communicates via one of several interfaces: SPI, I2C, UART, USB or analog/digital output
- Configuration utility runs on Windows, MacOS and Linux
Frequently bought together
Customers who bought this item also bought
Compare to similar items
This item Pixy (CMUcam5) Smart Vision Sensor - Object Tracking Camera for Arduino, Raspberry Pi, BeagleBone Black
|Shipping||FREE Shipping||FREE Shipping||FREE Shipping||FREE Shipping|
|Sold By||Charmed Labs||Charmed Labs||Charmed Labs||waveshare|
|Item Dimensions||1.5 x 2.25 x 2.25 in||3 x 2.9 x 1.4 in||1.5 x 2.25 x 2.25 in||3.94 x 7.09 x 3.54 in|
Pixy is a smart vision sensor you can quickly “teach” to find objects. It saves you time by only outputting the object data you're interested in. A multitude of connection options means you can use Pixy with almost any microcontroller. It connects directly to Arduino with the included cable, and fully supports Raspberry Pi and BeagleBone Black with included software libraries. Included in the box is mounting hardware to attach Pixy to your robot creation. The firmware, software and hardware are open source, so you can tweak to your heart's delight. Free tech support is included on the CMUcam wiki! Note: if you are using an Arduino shield in conjunction with Pixy, you may need to purchase Stackable Headers for the Pixy cable to fit. USB cable not included.
Seller Warranty Description90 day replacement warranty
Top Customer Reviews
I was expecting the sensor to work like my cell phone camera: providing crisp, clear colors that I could use to key off of for object detection. Instead, everything is sort of "washed out" making the brightest pink, nuclear orange, and sunspot yellow sticky notes all appear to have the same hue. Weird, but true. To help improve the "distance" at which it could see I tried using a narrow field of view lens: 12 degree. It was nice that I could simply unscrew the original lens and screw in the new one. The distance was great, about 8 feet, but this did not improve the detection capabilities despite having an IR cut-off filter.
In summary, it's a neat little device, but I would say it's still in beta and has lots of room for improvement before it will see wide adoption. For now, it's a niche sensor for some fun and interactive projects. Hopefully the improvements they make are purely software and my Pixy will get better over time, but I suspect that an improved sensor is in order which means my may quickly find itself holding down the corner of a drawer somewhere. Yes, it only does color blob detection. Not faces. Not shapes. Not anything greyscale (you cannot detect black, or white). So, if you want to do color blob detection (or face detection, or object detection), you can get the same result, and have a boatload more capability by using a Raspberry Pi with the Raspi Camera module for less money than the Pixy.
Secondly, although it works pretty well in demo situations where the lighting is constant, it performs poorly when the lighting conditions change. If you get it to recognize a color under one set of lighting conditions, it will no longer recognize it when the lighting shifts. This is a problem for us since we would like to use this on a robot which moves around which causes the lighting conditions to change.
Some of these problems might be fixable if we could modify the firmware, but although the firmware is open source, the tools required to build it are not and cost thousands of dollars. They have promised to provide open source tools, but, as far as I know, this promise has not yet been met.