![]() |
Whos doing stuff with the Kinect?
I wanted to start a tread for people to share what thier team is doing with the Kinect/planning to do.
Right now 2607 is just diving in, got some basic object detection. Time for more programming! |
Re: Whos doing stuff with the Kinect?
We did a lot of work with the Kinect last season through Beta Testing (http://team1912.com/beta_testing.html). We are looking at doing some more complex testing such as finer tuned movements and more realistic execution. We don't know if the Kinect will be a part of Beta Testing again but if it is I'd definitely like to revisit what we accomplished last year and step it up a level.
|
Re: Whos doing stuff with the Kinect?
We have the basic implementation coded but no test robot to practice it on.
|
Re: Whos doing stuff with the Kinect?
it's easy enough to write goal detection code using Microsoft's new SDK and EmguCV (as compared to the old SDK).
The problem is, should we find an onboard computer that can use Windows, and have to deal with slow startup/shutdown, or should we use freenect and openCV with linux... |
Re: Whos doing stuff with the Kinect?
Quote:
Not robotics related: I really want to work on object reconstruction based off the depth map readings after I circle the object with the kinect. Then be able to move the reconstructed image in the computer, be able to spin it, and what not. What i have done with the kinect is use it as a mouse. I had it detect my hand, then defined a region on the screen. The center of my hand in that region corresponded with where the mouse was on the screen. Then if the area of my hand decreased to a certain amount (I make a fist), then the mouse left clicked. Pretty simple stuff. |
Re: Whos doing stuff with the Kinect?
2485 had some great stuff going on in the pit and on their robot with the kinect.
|
Re: Whos doing stuff with the Kinect?
We are targeting with the kinect
|
Re: Whos doing stuff with the Kinect?
Quote:
In our pit at the Inland Empire regional, we use a Kinect and the Microsoft SDK to allow users to view and rotate the CAD model using their hands. The model is displayed on a large screen in the front of the pit. We translate hand movements into mouse input to rotate the model, which is displayed in SolidWorks. |
Re: Whos doing stuff with the Kinect?
We (2408) are targetting with the kinect! :D
|
Re: Whos doing stuff with the Kinect?
We're planning on using an ODROID-X2 (the fantastic version) to do on-board vision processing using OpenCV (but we're learning to use OpenCV on Windows in case one day we're allowed to have a laptop on-board). We have no idea what we're doing (it's difficult to install due to the use of things like "etc..." and "related software" mentioned on the OpenCV website under Linux/ARM installation instructions). On a non-robotics note, I seem to recall that we downloaded some Kinect software in my drafting class that allows you to export real time motion capture data straight into Autodesk MotionBuilder.
|
Re: Whos doing stuff with the Kinect?
1334 is working on a general Kinect project that runs linux/freenect. Currently, it's been 3 months since any work was done; I'm looking to continue it now that build season is over. If anyone would like to contribute to this project, the help would be appreciated.
|
Re: Whos doing stuff with the Kinect?
Quote:
What do you need help in, setting up the opencv library, or getting familiar with the functions and the process? Shoot me an email. I'd love to help. Our team's goal is to have every other regional team, all 7 of them, to use computer vision in some manner for the st louis regional next year. I think that'd be very cool. If anyone is interested in getting a workshop, email me. We are from the st louis region but will video tape the workshops. For a project to teach someone else opencv, I had him track the 3 different coloured disks, though we didn't have a pick up due to weight. We were informed of this after the fact. |
Re: Whos doing stuff with the Kinect?
The project is admittedly, slightly ambitious. I'm trying to create a flexible, reusable system that not only our team can use, but other FRC teams. I felt like several teams were re-inventing the wheel with Kinect vision, and, seeing no other similar project, started my own. I'm seeking input and contributers from other teams to hopefully make it as useful as possible. The repo is at https://github.com/frc1334/kinect-relay for those who are potentially interested. I'd call it about 40% done right now.
|
Re: Whos doing stuff with the Kinect?
Last year, we started trying to use the Kinect for animation, but things got very difficult very quickly, and we decided it wasn't worth our time. Maybe we were doing something wrong, but I'm not sure. Anyone else using it for animation?
|
Re: Whos doing stuff with the Kinect?
Team 2614 is using the Kinect for Vision Targeting this year! I will eventually get a white paper out describing the system in full. :]
|
| All times are GMT -5. The time now is 00:52. |
Powered by vBulletin® Version 3.6.4
Copyright ©2000 - 2017, Jelsoft Enterprises Ltd.
Copyright © Chief Delphi