Can we connect Microsoft's azure Kinect DK camera with Nvidia Jetson TX2 board DK together? Do these development kits support each other? Can we work on them?
Not yet, but we are working on it. Is should be ready soon though.
You can see our ARM proposal here: https://github.com/microsoft/Azure-Kinect-Sensor-SDK/blob/develop/proposals/ARM-support.md
Related
Does the Google Assistant SDK also work on a "Pi 2 Model B" (ARM Cortex-A7, a 32bit processor), or is a "Pi 3 Model B" ( ARM Cortex-A53, 64bit processor) essential to get the SDK up and running ?
It's working properly on a Raspberry Pi 2. I'm using it with my Logitech c920's mic.
I am running the Voice Kit that was in the MagPi Magazine on a Raspberry PI Zero W, so it should run fine on almost all the Pi's https://aiyprojects.withgoogle.com/voice/
I bought a dell 7559 laptop for deep learning. I got ubuntu 16.04 installed on it but I am having trouble getting caffe and tensorflow on it. The laptop used Nvidia Optimus technology to switch between gpu and cpu to save battery usage. I checked the bios to see if I can set it to use only gpu but there is no option for it. Using bumblebee or nvidia-prime didnt work either. I now have ubuntu 16 with mate desktop environment it is preventing from getting the black screen but didnt help with the cuda issue. I was able to install the drivers and cuda but when I build caffe and tensorflow they fail saying that it didnt detect a gpu. And I wasnt able to install opengl. I tried using several versions of nvidia drivers but it didnt help. Any help would be great. thanks.
I think Bumblebee can enable you to run Caffe/Tensorflow in GPU mode. More generally, it also allows you to run other CUDA programs on a laptop with Optimus technology .
When you have installed Bumblebee correctly (tutorial: Bumblebee Wiki for Ubuntu ), you can invoke the Caffe binary by pepending optirun before the caffe binary. So it goes like the following:
optirun ../../caffe-master/build/tools/caffe train --solver=solver.prototxt
This works for the NVidia DIGITS server as well:
optirun ./digits-devserver
In addition, Bumblebee also works on my dual-graphics desktop PC (Intel HD 4600 + GTX 750 Ti) as well. The display on my PC is driven by the Intel HD 4600 through the HDMI port on the motherboard. The NVidia GTX 750 Ti is only used for CUDA programs.
In fact, for my desktop PC, the "nvidia-prime" (it's actually invoked through the command line program prime-select) is used to choose the GPU that drives the desktop. I have the integrated GPU connect to the display with the HDMI port and the NVidia GPU through a DisplayPort. Currently, the DisplayPort is inactive. The display signal comes from the HDMI port.
As far as I understand, PRIME does so by modifying /etc/X11/Xorg.conf to make either the Intel integrated GPU or the NVidia GPU the current display adapter available to X. I think the PRIME settings only makes sense when both GPUs are connected to some display, which means there need not be an Optimus link between the two GPUs like in a laptop (or, for a laptop with a Mux such as Dell Precision M4600, the Optimus is disabled in BIOS).
More information about the Display Mux and Optimus may be found here: Using the NVIDIA Driver with Optimus Laptops
Hope this helps!
I'm now trying to set up my kinect version 2
I installed Microsoft sdk and usb extension board (Renseas 720202)
Question is, Kinect can be detected before driver of usb board install.
In Kinect verifier program, usb controller are not compatible but can get
color image with low fps.
ahttp://ppt.cc/XvKOg (Sorry I can't post more than two links. Plz remove 'a' and see the pic)
http://ppt.cc/kv38M
After install driver of usb board, I get nothing...
ahttp://ppt.cc/hbWAE
http://ppt.cc/Cw29u
I search that intel and Renaseas chips are compatible with Kinect v2.
What can I do? Thanks
update
In the end, I change computer which use intel chipset on mother board to solve the problem.
Could anyone get the camera data from the Kinect using a Raspberry Pi ?
We would like to make a wireless Kinect connecting it using Ethernet or WiFi. Otherwise, let me know if you have a working alternative.
To answer your question, yes it is possible to get Image and depth on the raspberry pi!
Here is how to.
If you want to use just video (color, not depth) there is already a driver in the kernel! You can load it like this:
modprobe videodev
modprobe gspca_main
modprobe gspca_kinect
You get a new /dev/videoX and can use it like any other webcam!
If you need depth (which is why you want a kinect), but have a kernel older than 3.17, you need another driver that can be found here: https://github.com/xxorde/librekinect. If you have 3.17 or newer, then librekinect functionality is enabled by toggling the gspca_kinect module's command-line depth_mode flag:
modprobe gspca_kinect depth_mode=1
Both work well on the current Raspbian.
If you can manage to plug your kinect camera to the raspberry Pi, install guvcview first to see if it does works.
sudo apt-get install guvcview
Then, typeguvcview in the terminal and it should open an option panel and the camera control view. If all of that does works and that you want to get the RAW data to do some image treatments, you will need to compile OpenCV (it takes 4 hour of compiling) and after that, you just will need to program whatever you want. To compile it, just search on Google, there are lots of tutorial.
Well, as far as I know there are no successful stories about getting images from Kinect on RaspberryPi.
On github there is an issue in libfreenect repository about such problem. In this comment user zarvox say that RPi haven't enough power to handle data from Kinect.
Personally I tried to connect Kinect with RPi using OpenNI2 and Sensor, but have no success. And that was not a clever decision because it's impossible to work with Microsoft Kinect on Linux using OpenNI2 due to licensing restrictions (Well, actually it is not so impossible. You can use OpenNI2-FreenectDriver + OpenNI2 on Linux to hookup Kinect. But anyway this workaround is not suitable for RaspberryPi, because OpenNI2-FreenectDriver uses libfreenect).
But anyway there are some good tutorials about how to connect ASUS Xtion Live Pro to RaspberryPi: one, two. And how to connect Kinect to more powerfull arm-based CubieBoard2: three.
If you intend to do robotics the simplest thing is to use the Kinect library on ROS Here
Oderwise you can try OpenKinect, They provide the libfreenect library that let you acess to the accelerometers the image & much more
OpenKinect on Github here
OpenKinect Wiki here
Here is a good exemple with code & all the details you need to connect to the Kinect & operate the motors using libfreenect.
You will need a powered USB hub to power the Kinect & to install libusb.
A second possiblity is to use the OpenNI library which provides a SDK to develop midleware libraries to interface to your application there is even an OpenNi lib for processing here.
yes, you can use Kinect with raspberry pi in a small robotic project.
I have done this work with the openkinect library.
my experience is you should check your raspberry pi and monitoring pi voltage, not time does to low voltage.
you should accuracy your coding to use lower processing and run your code faster.
because if your code had got a problem, your image processing would be the slower response to the objects.
https://github.com/OpenKinect/libfreenect https://github.com/OpenKinect/libfreenect/blob/master/wrappers/python/demo_cv2_threshold.py
Microsoft has recently released Kinect SDK 1.5 and some very neat associated features such as face tracking.
I have a Kinect sensor for XBox 360 and Windows 7 (driver, Kinect studio) do not seem to recognize the device.
Can anyone advise if this is an "operator error" or if SDK 1.5 indeed does not support Kinect for XBox sensor but only Kinect for Windows (I have USB and power adapter for it).
Thank you,
Edmon
As Chris Ortner pointed out, the Kinect for Xbox sensor is compatible with Windows 7 for development purposes only - you need to have the SDK installed.
It might be worth checking to see if the SDK has installed properly.
In Device Manager you should see if it has the following:
Microsoft.Kinect
Microsoft Kinect Audio Array Control
Microsoft Kinect Camera
Microsoft Kinect Security Control
And:
Sound, video and game controllers
Kinect USB Audio
I have read that it is recommended that you do not have the sensor connected when installing the SDK. This is also implied by the msdn article setting up a kinect sensor. I have also read, although unable to provide a reference, having things like OpenNI or libfreenect may cause issues and should be uninstalled before installing the SDK. The link also provides other troubleshooting tips.
One last thing to check, ensure you are plugged into a USB 2.0 or USB 3.0 bus. It is one of the hardware requirements, and I am aware that one of the KinectSensor states is InsufficientBandwidth.
The SDK supports Kinect for Xbox360 as a development device, so this is probably an operator- or hardware error.
Please keep in mind, however, that the SDK must be installed for the Xbox360 version to work - the device is not supported for deployment on machines which have just the driver installed.