How to Use Kflash_GUI to Burning Firmware for M5Stick V?
Download
Start
Hey there! To get started, just connect your device to your computer using the Type-C cable. Then, double-click to open the Kflash_GUI application. Here, you can select the corresponding device port, board type (M5StickV), firmware program, and baud rate. All you have to do is click Download and you're ready to go!
Classification Model
We're so excited to get started training the model! We just need to make sure we have lots of training material. Since the hardware on the M5StickV and UnitV are a little different, we also need to make sure we're shooting the material in the right way. Please take a look at the instructions below to make sure you're shooting according to the hardware you're using.
M5StickV-Shooting Footage
To get started shooting the training material, you'll first need to download the boot-M5StickV program zip file. Once you've downloaded it, simply unzip all the files in the zip file to the SD card (M5StickV requires an SD card, click here to see the support type).
Once you've inserted the SD card for storing picture material, you're ready to boot up the camera! Just long press the left power button to boot up the camera. When the screen appears, as shown in the figure below, you'll know you're in the shooting program.
You'll see a navigation bar at the top of the screen that shows you the current class number and the number of pictures you've taken in real time. Just press the HOME button to take pictures! You can also use the buttons on the right side of the camera body to switch the class number.
We're so excited to announce that the program currently offers 10 groups of Class for users to capture those training materials! Each group of Class represents a different kind of recognition object. To get the best training results, we recommend shooting more than 3 groups of Class. We just want to make sure we're giving you the best training experience, so each group of Class needs to be shot more than 35 pictures. Otherwise, it won't be available through the training in the cloud. Thanks so much for your understanding! The more clips you take, the better the recognition training effect and the higher the recognition rate!
UnitV-boot-UnitV
The Boot-UnitV program is here to help you capture images for UnitV. It's a handy little tool that collects material for model training.
To get started, simply insert the SD card into UnitV's card slot. That's where the recorded training material will be stored by default.
We're always here to support you! If you need any help along the way, don't hesitate to reach out. (Just a heads-up: UnitV needs an SD card. You can see which types are supported here.)
To take a picture with the IDE's camera, just press the A button to take a picture and the B button to switch the class number. Easy-peasy! The output log is there to help you out! It shows you the class number for each operation and the number of pictures taken. We're so excited for you to see the log output! Just click "Serial Monitor" below to get started.
We're so excited to announce that our program currently offers 10 groups of classes for users to take training material! Each group of classes represents a different kind of recognition object. We really encourage you to get the best training results you can! To do this, we suggest you try shooting more than three groups of class. We really want to make sure that everything is as accurate as it can be, so we just ask that you try to shoot more than 35 pieces of material for each group of Class. Thanks so much for your help with this! The more clips you shoot, the better the training effect and the higher the recognition rate!
When you're shooting your training materials, we'd love it if you could try to keep the lighting as consistent as possible. This will help us to get the best recognition results! Also, try to fill the screen with your recognized object, and make sure there are no distracting elements in the background.
Just a heads-up! To make sure everything goes smoothly, each group of classes needs to have at least 35 clips. Otherwise, it won't be able to be sent for cloud training. Thanks! The more clips you shoot, the better the recognition training will be!
