Let's all meet at the HackerDojo for a weekend and make some awesome Kinect hacks together!
------
Although we're planning a couple of workshops, this is definitely more of a social brainstorming and coding event than a thorough introduction to the Kinect for beginners.
------
1. Being prepared:
If you want to do software development on the Kinect and have no prior experience on the Kinect, you should do the following two things before showing up:
First, I strongly recommend purchasing a Kinect prior to attending. Having your own Kinect makes Kinect hacking way more fun than working with canned data. While we may have a spare or two availble, there's no guarantee of thtis and you may have to share it with a large number of others.
Second, I strongly recommend getting the drivers running on your computer and familiarizing yourself with the OpenNI programming environment or equivalent prior to the hackathon. Your hacks will be way cooler you are ready to go instead of struggling with driver issues.
Some good resources for getting started and installing the drivers:
Option 1: The OpenNI and NITE driver installation -- If you want your hack to involve body or gesture tracking, this is the way to go. https://github.com/avin2/SensorKinect (see Install Notes)
If you encounter any trouble, first read the archives of the OpenNI mailing list. If you ask me a question about OpenNI, I'll probably just go there to look it up. http://groups.google.com/group/openni-dev
Option 2: ROS and PCL driver installation -- It's good for doing heavy point cloud manipulation and object detection, but has a steep learning curve and only runs smoothly on Ubuntu Linux.
http://www.ros.org/wiki/ROS/Installation (first)
http://www.ros.org/wiki/openni_kinect (second)
We will try to help out whenever possible but please don't come expecting us to solve all of your problems.
If you are well versed in OpenNI or ROS and want to serve as an expert assistant to newbies, please let me know.
---
2. Schedule:
Saturday:
10am -- Room becomes available
10:30am -- We will have a quick go-round where we talk about what we're interested in hacking. This will help potential groups form.
11am-4pm - Open time for workshops. We will have the upstairs room reserved for anyone who wants to run a workshop.
10am-10pm - General hacking time
Meals will be organized ad-hoc, with people contributing to the event.
Sunday:
10am-4pm - General hacking time
4pm-5pm - Presentation of the weekend's hacks.
----
3. Workshops planned:
- Amir Hirsch will be doing a workshop on using the Unity Engine to make 3D games with the Kinect.
- If there is enough interest, I will do a workshop on using ROS and PCL to manipulate Kinect point clouds. This has lots of uses for object tracking and recognition. If at least 5 people email me ahead of time saying they want this to happen, I'll do it. :-)
- I'd *really* like it if someone other than me ran an OpenNI/NITE introduction workshop. I've done some OpenNI/NITE hacking on Windows, but it would be good to have someone who's more of an expert give the introduction.
If you'd like to do another workshop, please let me know ahead of time. Workshops should be 30-60min long and be designed to
----
4. Logistics:
Where: Hacker Dojo, 140 South Whisman Road, Mountain View, CA
http://goo.gl/maps/kzea
Sign in at the front desk. Walk to either side of the desk to get through to the main area. Look for the Kinect Hackathon sign. There will be a greeters' desk with basic information.
If you are driving to the Dojo it is VERY IMPORTANT that you only park in Dojo parking spaces:
See: http://wiki.hackerdojo.com/w/page/25425/Dir...
For directions, See: http://wiki.hackerdojo.com/w/page/25425/Dir...
If you are walking CalTrain to Mountain View, the Dojo is a 1mi flat walk from the CalTrain train station. Or you could transfer to San Jose light rail for one stop which will make it 0.4mi -- but it will probably take longer and cost more! Walking directions from either station are on the directions page above, and have a slight short cut that Google Maps doesn't find.
Comments