With regards to the HUSKYLEN - i have the following questions:1. Does it have an SDK to allow integration? 2. Does it come with a protective casing?3. Is there any documentation?4....

Can I use huskylens with another display or at least longer cable camera so the camera is more in selfie mode to be used when looking at the screen and camera facing you?

Hey had a question about the display, Is there a way to make what the camera sees to another device/screen?

Do you have a cable to connect Huskylens with Grove port?

Thank you very much for your feedback, but due to the technical limitations of the hardware, we are unable to fulfill this requirement for the time being.
We will forward this request to our product department and hope to improve this issue in subsequent products.

I am looking for weld seam tracking using

Gravity: Huskylens - An Easy-to-use AI Camera | Vision Sensor is it possible to communicate with aurdino to drive the robot in diffrent directions?

This really is not fit for purpose, I wouldnt recommend this as being part of any real world solution. You can play around with it, but therre is not enough documentation, the device works intermittantly (even for simple tasks), and constantly needs rebooting when stuck or crashes. Use a Pi camera, connected to a Pi Zero W 2, Pi 2, Pi 4 etc for a cheaper, more reliable system. All of teh above tasks can be done using python and opencv.

Hi, Qr code recognition is a function of Huskylens Pro, which is currently not available online.

Someone might need huskylens' camera connection diagram：
https://uploads.disquscdn.c...

After training an object tracking network with Learn and Auto Save both enabled, I then disable learning. Is the Huskylens supposed to still draw a yellow rectangle on the screen when not in training mode, as its looking for the object ? Or does it not draw the yellow rectangle unless its in learn enabled mode ?

I dont think you'd want to pass the image to an Arduino, but the Huskey Lens has features that can find/identify/track an image, and you can send that output ie location, object ID to the Arduino. The comms between the Huskey and Arduino (or any other board) can be UART or I2C. Both require 2 cables, Arduino supports both in hardware.

Hello. What about the hardware schematics? I need to replace the display that I accidently broke.

Tip: for "Object Recognition" mode: while there are 20 or so predefined objects, there is a bit of a "trick" needed to associate an ID number with each recognized object when monitoring the serial data. At first, serial data seemed useless, as it was always set to "0" regardless if the object being detected was "person", "dog", "car", etc... However, you need to enable multiple object learning (a hidden menu - see wiki doc). Then, you'll learn some of those 20 objects into respective ID numbers. "Learn" is a bit misleading, as the object recognition is already learned (built-in models), like "dog". So to give dog an actual ID num other than 0, you point the HuskyLens at a dog picture, then hit learn button, dog gets ID 1, hit learn again, now point at a Person picture, hit learn, person get ID 2, hit learn, point at a car pic, car gets ID 3 ..... and you keep going (if you want). I just did those three and stopped. Again, you are not "learning" a new dog, or specific dog, hence why I used stock photos, you are just getting Husky to detect dog on-screen, so you can assign a ID number to it with the learn button press. It's a really weird methodology, but works. Now in my Python code, I just hard-code 1 2 3 to the respective objects to make my project take action based on what get detected.

hello, when compiling the sketch, errors appear in these lines of code: DFRobot_Servo.h, huskylens.writeOSD, writeAlgorithm..., arduino says it does not recognize these parameters, what is the solution, where are they these bookstores? You have this sketch on your page, so, I think, you should give us free access to these bookstores. Thanks.

The camera of Huskylens has 200W pixels. The recognition distance depends on the size of the object. As long as the object is clearer on the screen, it can be recognized.

Hello!
I wondering if it's possible to disconnect the camera lens without damaging any other components, or compromising it's functionality (I've been working on a robot that only has the camera visible, with the screen hidden). Thanks!

How can I train my own model? I have an image dataset and I want to use it instead of taking pictures by hand

How can i find the number of people within a room with Huskylens + Arduino? Is it feasable?

Hello Vijay,

We currently have no device that has built-in algorithms for Driver drowsiness monitoring and Driver behavior.

Hi.
Let's say I have two objects stored in the Husky: one painted red, the other green. Is there any way for it to count in numbers how many times these two objects crossed and store this data in a photo format? (Vector motion detection)

https://uploads.disquscdn.c...

Hi.
Let's say I have two objects stored in the Husky: one painted red, the other green. Is there any way for him to count in numbers how many times these two objects crossed and store this data in photo format?

https://uploads.disquscdn.c...